51 results about "Function pointer" patented technology

Techniques for identifying malware based on system API function pointers are disclosed. In some embodiments, a system / process / computer program product for identifying malware based on system API function pointers includes monitoring changes in memory during execution of a malware sample in a computing environment; detecting a dynamic evasion behavior using an Application Programming Interface (API) vector comprising a plurality of system API function pointers identified in the memory during execution of the malware sample in the computing environment; and generating a signature based on the API vector for automatically detecting the malware during execution in the memory, wherein the malware sample was determined to be malicious.

The invention discloses a function processing method and device, equipment and a storage medium, and relates to the technical field of computers. The method comprises the following steps: sending an interception request through a Java layer, wherein the interception request is used for indicating a first function and a second function; the first function is an intercepted party for function interception, and the second function is an intercepting party for function interception; receiving an interception request through an operating system bottom layer; and modifying a function pointer of the first function into a first function pointer through a bottom layer of the operating system, the first function pointer pointing to the second function, so that the second function intercepts a call request for the first function. According to the method, any function can be intercepted under the condition that the complexity of the original code of the application program or the intercepted function is not increased, so that the application range and scene of function interception are expanded.

The invention provides a method and a system for dynamically calling a kernel unexported function, which are used for dynamically acquiring a function address of an unexported kernel function and forcibly calling the function through a kProbes mechanism, matching with a function pointer type definition and a mechanism for forcibly converting the function address into a function pointer. The method not only supports the calling of the kernel function which is not exported, but also does not need to carry out hard coding on the address of the called kernel function, supports the dynamic acquisition of the address of the kernel function, avoids the change of the coincidence address caused by the difference of kernel versions, greatly enhances the compatibility of the software package of the kernel function which is not exported, and improves the reliability of the software package. And the packet does not need to be recompiled in each kernel small version, so that the problem that the kernel symbol table is not matched and the kernel module cannot be loaded is avoided.

A system includes a memory and a processor in communication with the memory. The processor is configured to supply a library with a list of safe callback values, protect the list of safe callback values, invoke a callback, and validate the callback against the list of safe callback values to determine a status of the callback. The status of the callback is one of safe and unsafe. Additionally, the processor is configured to execute the callback responsive to determining the status of the callback is safe. The processor is also configured to abort the callback responsive to determining the status of the callback is unsafe.

The invention relates to a graph rendering method and device, computer equipment, a readable storage medium and a product. The method comprises the steps of calling an interface function driven by a first DRI in response to a graphic rendering request; wherein the first interface function is used for returning to a structural body, the structural body comprises a function pointer of a first DRI drive and a function pointer of a second DRI drive, and at least one function of the second DRI drive is different from the function of the first DRI drive; calling a first rendering function based on a function pointer driven by the first DRI; and in the process of processing the graphic rendering request through the first rendering function, if the graphic rendering request relates to a first function in the at least one function, calling a second rendering function based on a function pointer driven by a second DRI to perform graphic rendering. By adopting the method, the graphic rendering effect can be improved.

The invention provides a GPU instruction optimization method and device, equipment and a storage medium, and the method comprises the steps: initializing an instruction set and a hash table in an NIR optimization stage; initializing a function pointer, and obtaining a current function according to the function pointer; initializing a basic block pointer according to the current function, and obtaining a current basic block according to the basic block pointer; initializing an NIR instruction pointer according to the current basic block, and obtaining a current instruction according to the NIR instruction pointer; according to whether the current instruction is a jump instruction, adding the current instruction into an instruction set and a hash table; and in an emit assembly instruction stage, performing instruction optimization according to the instruction set and the hash table. According to the method, in the emit assembly instruction stage, instruction optimization is carried out according to the instruction set and the hash table, the number of NIR and the number of assembly instructions are reduced, and then the register pressure is reduced.

The application discloses a non-restart code repair method using memory gaps, comprising the following steps: screening target objects of repair codes; searching memory layout, managing and storing information of retrieved memory gaps, and matching space information corresponding to the repair codes; generating a code repair scheme by comparing modification requirements of functions in the repair codes, replacing modified function pointers or parameter structures of the functions; updating information of a kernel symbol table; writing data and data structures required to be repaired by the target objects into the memory gaps, generating a kernel module according to the code repair scheme, and realizing code repair of a non-restart system by loading the kernel module and using a HOOK mechanism. The application is suitable for different types of code repair patches in real environments and has good universality.

The present invention discloses a method and apparatus for identifying and protecting function pointers and their data dependencies. The method first identifies memory objects belonging to function pointers or containing at least one function pointer from a given program, then uses taint analysis to identify function pointers and their data dependencies, and then uses Intel MPK to protect this data. This ensures their integrity, thereby preventing the execution path of the program from being modified and improving the security of the program.

The embodiment of the invention discloses a Linux kernel cross-version driving implementation method and device and a medium, and the method comprises the steps: carrying out the instantiation of a function pointer in a newly-added structural body through employing a newly-added instrumentation eBPF program in a device driving frame in a kernel during the initialization of a device driver, so as to point to a function in the structural body, and carrying out the instantiation of the function pointer in the structural body; the function in the structural body is a drive function of a new version realized in an eBPF program; a user program accesses the device through system calling, when a kernel calls the old-version driving function in the file opportunions structural body of the general character type device driving framework, the eBPF program is used for carrying out hook association on the old-version driving function and the corresponding new-version driving function, the old-version driving function is intercepted, and the new-version driving function is intercepted; and turning to execute the driving function of the new version. And the problem of compatibility of the kernel with the driving module in kernel version evolution is avoided.

The present application relates to a kind of executable engineering dynamic loading and debugging method based on RISC-V, based on the original engineering containing static library generates main engineering and independent executable engineering;Increase the dynamic call running interface function written with assembly language in main engineering, so that main engineering can load the executable file generated by executable engineering;Executable engineering establishes a function pointer table for calling main engineering function, realizes the call to main engineering function;The main engineering and independent executable engineering are dynamically debugged respectively;In debugging engineering, create adaptive processing function of debugging print information, so that program can automatically identify current environment, such as in debugging environment, then carry out normal output debugging information, such as in non-debugging environment, program can skip output debugging information interrupt instruction and continue running. Realize the flexible inter-module function call mode based on RISC-V processor and realize the environment adaptive processing function of debugging print information.

The invention belongs to the technical field of embedded software testing, and provides a VSIPL vector library function testing method and device.The method comprises the steps that operator functions to be tested are classified and abstracted into uniform function pointers, test cases are constructed based on the function pointers, and a corresponding relation table of the test cases and VSIPL data object specifications is established; automatically generating a matched VSIPL data object according to the relation table, calling the operator function to be tested and the reference operator function corresponding to the operator function to perform calculation on the same data, outputting a test result based on a calculation result and consumed time, and releasing resources. According to the method, on-demand self-adaptive generation of test data and unified scheduling of operator functions are realized through a cooperation mechanism of a function pointer and a relation table, the automation degree, coverage integrity and cross-platform verification efficiency of VSIPL vector library test are remarkably improved, batch performance and precision evaluation is supported, and the method is suitable for popularization and application. And the correctness and the acceleration effect of the optimized version function are effectively ensured.

The invention discloses a quantum bit calibration method and device and a storage medium, and the method comprises the steps: obtaining a universal function structure body corresponding to a to-be-called calibration function for the to-be-called calibration function in a quantum bit calibration graph; according to the general function structural body, constructing a general function prototype structural body corresponding to the to-be-called calibration function; wherein the general function prototype structural body is provided with a function pointer, and the function pointer points to a to-be-called calibration function; according to the general function prototype structural body, constructing an actual call calibration function corresponding to the to-be-called calibration function; the actual call calibration function is called to complete quantum bit calibration based on the quantum bit calibration graph, and when the actual call calibration function is called, the actual call calibration function calls the calibration function to be called through the function pointer to operate the function body of the calibration function to be called. According to the quantum bit calibration method and the quantum bit calibration device, automatic quantum bit calibration can be realized by using an original manually written calibration function under the condition of relatively low transformation cost.

A program running method and a device are described for improving efficiency of running an application. The method includes obtaining a dynamic image of an application, where the dynamic image includes a trampoline function pointer segment, the trampoline function pointer segment includes at least one trampoline function pointer, the at least one trampoline function pointer corresponds to at least one interface, and the at least one interface includes an interface of a kernel or an interface of a common code library The method also includes calling the interface of the kernel or the interface of the common code library based on the trampoline function pointer segment, to run the application.

The invention discloses a cross-platform Matlab real-time calculation method, which comprises the following steps of: 1) defining a function signature of a matlab algorithm function under a c / c + + language specification and a data structure of interaction between the function signature and an application program; 2) defining a function pointer, and initializing the function pointer to be null; 3) writing codes of the application program; 4) writing a script program of a matlab algorithm function; 5) converting a script program of the matlab algorithm function into a c / c + + code; 6) modifying a function pointer to point to an entry function address of the c / c + + code; 7) copying the c / c + + code and the code of the application program to a new directory, scanning all source files in the new directory, and compiling the source files into an executable program in the target running environment; 8, the executable program is transmitted to the target operation environment to be executable. According to the calculation method, the strong calculation capacity of Matlab is achieved, real-time data processing is achieved, the data processing performance is greatly improved, and the popularization value is high.

The embodiment of the invention discloses a data processing method and device, electronic equipment and a storage medium, the data processing method and device are applied to a Wasm virtual machine, and the method comprises the steps that in the running process of a Wasm program, function calling information of a native function is obtained; wherein the function call information comprises function index information and function parameter information; determining a target function signature corresponding to the function index information by using a general function in response to a call instruction for the general function detected in the operation process of the wasm program; wherein the target function signature comprises a function pointer of a target native function to be called; and calling the target native function pointed by the function pointer by utilizing the general function according to the target function signature and the function parameter information. By means of the method, the target native function can be simply and rapidly called, and the program running performance is good.

The embodiment of the invention provides a model application method and device, a vehicle, a medium and a program product. The method comprises the following steps: firstly, determining a current network layer corresponding to a current data input address according to a model structure array of a target model and the current data input address; then, according to a function corresponding to the function pointer of the current network layer and a weight corresponding to the weight pointer, calculating data in the current data input address, and generating a calculation result; then, storing a calculation result to a data output address corresponding to a data output address pointer of the current network layer; and finally, determining the data output address as a new current data input address, repeating the previous steps until the data output address is a model result output address, and determining the current calculation result as a model output result. Through the method, the flexibility of the model in the execution process and the later maintenance process is improved.

The present invention provides a method, device, ECU and storage medium for flashing a vehicle ECU. The method includes: generating the source code of the ECU firmware, compiling the source code of the ECU firmware to generate the ECU firmware, and performing flashing using the ECU firmware; wherein, the flashing includes: obtaining the converted binary machine code corresponding to the erase and / or write function from the ECU firmware; performing an inverse operation process on the converted binary machine code to restore it to the original binary machine code; storing the original binary machine code in an array of the same data type and size in the RAM, and assigning the array name to a function pointer of the corresponding type; calling the function pointer to perform erase and write operations, and after the call is completed, clearing the array in the RAM. By using the above technical solution, it is possible to directly include the flashdrv executable machine code data in the firmware while ensuring safety, greatly improving the flashing efficiency of the ECU.

An embodiment of the present invention discloses a method, device, and medium for implementing cross-version drivers in the Linux kernel. The method includes: when initializing a device driver, instantiating function pointers in a newly added structure using a newly added staking eBPF program in the device driver framework in the kernel to point to functions in the structure, where the functions in the structure are new version driver functions implemented in the eBPF program; when a user program accesses the device through a system call and the kernel calls an old version driver function in the file_operations structure of the general character device driver framework, using the eBPF program to perform hook association between the old version driver function and the corresponding new version driver function, so as to intercept the old version driver function and redirect to execute the new version driver function. This avoids compatibility issues of the kernel with respect to driver modules during kernel version evolution.

The invention provides a task processing method which comprises the following steps: when a target program initiates a call request for a graphic rendering unit, calling a target dynamic library to intercept the call request, and querying a function pointer table by using the call request under the condition that the function pointer table is initialized, the driving manager receives a query result sent by the computing device, transmits the call request to a driving program corresponding to the call request according to the query result, and returns an execution result to the target program, and when the computing device starts or switches the driving program of the graphics processor, the function pointer table is initialized by the driving manager. By means of the mechanism, the switching requirement between the driving programs of different graphics processors can be met, and the purpose of guaranteeing normal operation of the driving programs of the graphics processors in the task processing process is achieved. Meanwhile, the method can also support the calling requirement of the graphics processor when the central processing unit is used as the virtual graphics processor in the computing equipment.

The invention relates to the technical field of neural network middleware underlying scheduling, and discloses a neural network adaptive architecture system for business intelligent decision making. Comprising a feature extraction module for extracting a dimension feature identifier of an operator to be executed, a kernel mapping engine for matching a target operator kernel according to the dimension feature identifier, a scheduling execution unit for modifying a bottom layer function pointer according to the target operator kernel and executing a chain type entry pre-binding regulation, and a control module for controlling the scheduling execution unit to execute the chain type entry pre-binding regulation. According to the method, logic jump between operators is converted into hardware-level addressing through space-time overlapping of a calculation period and an addressing period, the scheduling arbitration overhead at the operator boundary is eliminated, the zero pause feeling of an instruction supply end is guaranteed, and the instruction sequence is pushed to the instruction high-speed cache through the address bus idle period when prediction fails. And the energy efficiency ratio and response certainty limit of the system are improved.

This invention relates to the field of encryption machines and discloses a method, system, device, and medium for algorithm optimization. The method includes: creating a mutex lock and a function pointer to an algorithm, and creating corresponding threads based on different implementations of the algorithm; starting the threads and executing the algorithm implementation corresponding to each thread; preempting the mutex lock according to the execution sequence; obtaining the thread that has locked the mutex lock and is pointed to by the function pointer, thereby obtaining the optimal implementation of the algorithm. This disclosed method facilitates the selection of the optimal algorithm in the current execution environment, and therefore allows for dynamic selection of the algorithm implementation according to actual needs, making it more flexible in use and operation.

The application discloses an RTOS task hook type adaptation method based on a compiling period static trampoline function table, and comprises the following steps: S1, a compiling period presetting stage; S2, a running period binding stage; S3, a running period calling stage; and S4, a running period canceling stage. According to the application, a static trampoline function is pre-generated in the compiling period, all codes are solidified to a read-only Flash area after being compiled, and any code does not need to be generated in the running time, so that the contradiction between the running time code generation and the MCU hardware characteristics is fundamentally solved, the system complexity and the Flash occupation are reduced, a one-to-one mapping relationship between a user hook function and the trampoline function is established through a bidirectional binding table, and when being canceled, the corresponding slot can be reversely positioned only through the user hook function pointer, the whole hook list does not need to be traversed, the time complexity is O (n), the real-time loss of a traditional dynamic link table scheme is eliminated, and the resource leakage and the hook residual problem are avoided.

The invention provides a function calling method, electronic equipment, a storage medium and a program product. Before simulation of a to-be-tested object is carried out, in the compiling process of the to-be-tested object, statistics is directly carried out on related compiling data of function data contained in the to-be-tested object, and a query information table is formed. Afterwards, during simulation, query results such as a function pointer and a return value offset address of the corresponding function data can be directly determined from the query information table according to a target object pointed by current simulation of the object to be tested and related information of a simulation instruction, and finally, specific calling of the function is performed according to the query results. In this way, in the application process, the incidence relation between the corresponding calling module and the to-be-tested object is decoupled, the compiling performance of the simulator is improved, and the memory space consumed in the simulation stage is reduced. And corresponding code utilization and code maintainability are improved regardless of segmentation compilation and non-segmentation compilation.

The present disclosure relates to the field of software technology, and specifically to a method, apparatus, device, and medium for calling an external script. The method comprises: reading an external script file and obtaining an assembly based on the external script file; obtaining a function pointer corresponding to any method in the assembly, and obtaining entity indication information of an entity acted upon by any method; creating a first key-value pair container with the entity indication information of the entity acted upon by any method as a key and the function pointer corresponding to any method as a value; obtaining any entity type in the external script file, and determining entity indication information of at least one entity corresponding to the entity type based on the entity type; obtaining a target function pointer from the first key-value pair container with the entity indication information of at least one entity as a key, and executing the function indicated by the target function pointer. This solution can increase the speed of calling an external script, reduce the probability of delays caused by calling an external script, and thus improve the user experience.

The application relates to the technical field of process optimization, and provides an SDK calling method and device, computer equipment and a storage medium. The method comprises the following steps: receiving a processing request carrying a function parameter; querying SDK information of a preset third-party application according to the processing request to obtain a third-party application function class corresponding to the processing request, performing a non-empty judgment on the third-party application function class to obtain a judgment result; when the judgment result is non-empty, obtaining a function function pointer address of the third-party application corresponding to the processing request according to the processing request; obtaining an initialization function function according to the third-party application function class and the function function pointer address; assigning the initialization function function according to the function parameter to obtain a target function function; and calling the target function function to process the processing request. The method can enable a small program SDK to normally process a processing request.

The application relates to a RISC-V-based dynamic loading program method, which converts a static library in an original project into an independent executable project, and comprises the following steps: modifying a link configuration file of the executable project to specify an executable project code address; modifying an executable project environment initialization process code, so that the executable project has the capability of returning to the original project, and the executable project register and the value stored in the general register of the original project are isolated from each other; the original project removes a static library file, and modifies a static library calling mode in the original project according to a static library function name into a dynamic calling executable project mode according to a function pointer; and a dynamic calling running interface function written in an assembly language is added to the original project, so that the original project can call an executable file generated by the executable project. The application provides a high-efficiency and stable dynamic loading program method under a RISC-V CPU environment without MMU.

The invention relates to the technical field of software, in particular to an external script calling method and device, equipment and a medium, and the method comprises the steps that an external script file is read, and a program set is acquired according to the external script file; obtaining a function pointer corresponding to any method in the program set, and obtaining entity indication information of an entity acted by any method; creating a first key value pair container which takes entity indication information of an entity acted by any method as a key and takes a function pointer corresponding to any method as a value; obtaining any entity type in the external script file, and determining entity indication information of at least one entity corresponding to the entity type based on the any entity type; and obtaining a target function pointer from the first key value pair container taking the entity indication information of the at least one entity as a key, and executing a function indicated by the target function pointer. According to the scheme, the speed of calling the external script can be increased, and the probability of delay caused by calling the external script is reduced, so that the user experience is improved.

Embodiments of the present invention disclose a method, apparatus, computer device, and storage medium for protecting sensitive code. The method for protecting sensitive code includes: compiling the obtained sensitive code into a sensitive file and extracting sensitive machine instructions in the sensitive file; embedding the sensitive machine instructions into non-sensitive code; defining a call function pointer of the sensitive machine instructions in the non-sensitive code to call the sensitive machine instructions in the non-sensitive code. During the process of maintaining sensitive code, non-public sensitive code can be directly added without modifying the link control file that links the sensitive file and the non-sensitive file, simplifying the steps of generating an executable file. In addition, while maintaining sensitive code, non-sensitive code can also be synchronously maintained, avoiding the risk that the generated executable file cannot run reliably due to incompatibility between the maintained sensitive file and non-sensitive file.

Using a common reference address when processing calls between a native application binary interface (ABI) and a foreign ABI. Based on a caller calling using a reference address, a lookup structure is used to determine whether the reference address is within a memory range storing native code and that the callee is native, or a memory range not storing native code and that the callee is foreign. Execution of a callee is initiated based on one of (i) calling the callee using the reference address within an emulator when the caller is native and the callee is foreign; (ii) calling an entry thunk when the caller is foreign and the callee is native; (iii) calling an exit thunk when the caller is native and the callee is foreign; or (iv) directly calling the callee using the reference address when the caller is native and the callee is native.

The invention provides a GPU instruction optimization method and device, equipment and a storage medium, and the method comprises the steps: initializing an instruction set and a hash table in an NIR optimization stage; initializing a function pointer, and obtaining a current function according to the function pointer; initializing a basic block pointer according to the current function, and obtaining a current basic block according to the basic block pointer; initializing an NIR instruction pointer according to the current basic block, and obtaining a current instruction according to the NIR instruction pointer; according to whether the current instruction is a TEXKILL instruction, adding the current instruction into an instruction set and a hash table; and in an emit assembly instruction stage, performing instruction optimization according to the instruction set and the hash table. According to the method, in the emit assembly instruction stage, instruction optimization is carried out according to the instruction set and the hash table, the number of NIR and the number of assembly instructions are reduced, and then the register pressure is reduced.