11 results about "Dead code" patented technology

The present invention relates to a method performed by a system (110) operable to configure a chemical separation system (120), the method comprising: receiving system hardware configuration data, wherein the system hardware configuration data is at least indicative of fluid manipulation modules (121_1-12N_M) of the chemical separation system (120), desired functionality of the fluid manipulation modules, and a fluid network (210) comprising fluid couplers configured to fluidly couple the fluid manipulation modules (121_1-12N_M), identifying a first set of function blocks to provide the indicated fluid manipulation modules with the desired functionality, generating a second set of function blocks by purging a third set of function blocks from the first set of function blocks to eliminate dead code using the configuration data, sending the second set of function blocks to the system (120).

The embodiment of the application provides a code processing method, device, computer equipment, computer readable storage medium and computer program product, and belongs to the technical field of software development. The code processing method comprises the following steps: before performing a target change operation on a target code library, scanning the target code library by using a dead code detection tool to obtain a baseline dead code set; in the case that the target change operation on the target code library is completed, scanning the target code library by using the dead code detection tool to obtain a current dead code set, the storage structure of the current dead code set being consistent with the storage structure of the baseline dead code set; determining the difference between the current dead code set and the baseline dead code set, and determining the incremental dead code of the operation based on the difference; and performing invalid code cleaning based on the incremental dead code. The technical scheme of the embodiment of the application can effectively identify the newly added dead code, and improve the efficiency of dead code cleaning.

The present application relates to the field of compilation. Provided are a dead code optimization method for a compiler, and a compiler, a processor and an electronic device. The dead code optimization method for a compiler comprises: acquiring a dead input parameter from among input parameters of codes to be optimized; on the basis of the dead input parameter, determining a dead code from among the codes to be optimized; and optimizing the dead code, so as to reduce the number of machine codes generated from the compilation of the optimized code by a compiler. The dead code optimization method for a compiler can simplify a program code and improve a code execution rate.

The invention relates to a depth code model-oriented hybrid backdoor attack method, which comprises the following steps of: inputting a code into a hybrid trigger, obtaining the code inserted into the hybrid trigger through a hybrid trigger function, then inputting the code into a polluted model, and outputting a preset target result by the polluted model; the contaminated model is obtained by randomly selecting part of samples from an original data set as targets and inserting a hybrid trigger, for each selected sample, generating a new tag by a hybrid trigger function, the hybrid trigger function comprising a function signature trigger and a dead code trigger, adding the modified samples into the original data set to obtain a polluted data set, wherein the polluted data set comprises unmodified samples in the original data set and the samples inserted into the hybrid trigger; and training the model by using the polluted data set to obtain a polluted model.

The invention discloses an FPGA (Field Programmable Gate Array)-based parallel compiling method and system and a medium, mainly relates to the technical field of parallel compiling, and is used for solving the problems that the utilization rate of a DSP (Digital Signal Processor) in an existing scheme is reduced, and an existing high-level comprehensive tool lacks automatic vectorization capability and cannot bind operation to a single DSP. Comprising the steps that a preset high-level language is converted into LLVM IR through the front end of an LLVM compiler; screening out a plurality of binary operation instructions meeting a candidate instruction identification rule from the LLVM IR; combining the binary operation instructions into tuples meeting the digit width of the input port of the DSP in a dynamic grouping rule; based on an instruction optimization rule, packaging the tuple into a final operation instruction for the DSP, and combining operands of the instructions in the tuple into an operand of the final operation instruction; and inserting the final operation instruction into a preset replacement point through a tuple replacement rule, and marking an original instruction tuple of the preset replacement point as a dead code.

This disclosure describes a computer-readable medium storing instructions for detecting malicious activity by monitoring the execution of intentionally weak code paths embedded within executable code. A monitoring system receives a mapping that associates each injected weak path with a unique identifier, then observes the executable during runtime to determine whether any such abnormal path is taken. Because these paths are designed not to execute during normal operation (implemented through inhibited or dead code reachable only under adversarial manipulation) their activation serves as a signal of potential attack behavior. When the monitoring system detects execution of a mapped intentionally weak code path, it generates a notification within a secure environment, enabling rapid identification of suspicious or unauthorized activity. This technique enhances software security by providing a lightweight, observable mechanism for detecting deviations from expected execution flows.

The invention discloses an invalid code identification method, which comprises the following steps: when a client detects that a certain micro-service is started, acquiring related information of the micro-service, such as a name of the micro-service and a branch to which the micro-service belongs, and sending the related information to a server; and the server obtains a source file and / or a compiled file corresponding to the micro-service based on the related information, and sends the source file and / or the compiled file to the client. After a client receives a source file and / or a compiled file and starts to run micro-service, the running condition of codes in the source file and / or the compiled file is detected through a probe, and therefore invalid codes are determined. Namely, by introducing the probe, the running condition of the code in the micro-service can be monitored in real time, so that the invalid code in the micro-service can be accurately identified, and the reliability and safety of code cleaning are improved.

The present invention provides an invalid code optimization method for a compiler, a processor, and an electronic device, relating to the field of compilation. The invalid code optimization method for a compiler includes: obtaining an invalid input parameter from an input parameter of a to-be-optimized code; determining an invalid code from the to-be-optimized code based on the invalid input parameter; and optimizing the invalid code to reduce the number of machine codes generated by the compiler when compiling the optimized code. The above invalid code optimization method for a compiler can simplify program code and improve code execution efficiency.

This invention discloses a robustness enhancement method for large-scale code models based on context processing and post-training, selecting the enhancement method according to the type of model to be enhanced. For post-training enhancement, adversarial perturbation samples are first generated from the given source code using equivalence semantic transformation, greedy search, and simulated annealing. These perturbation samples are then input into the post-training framework to fine-tune the target large-scale language model. For context enhancement, a backup dataset is generated using backup code. The optimal subset of examples most semantically similar to the current problem is selected by calculating similarity. Dead code is then normalized, and the optimal subset of examples is adaptively selected. Contextual prompts are dynamically adjusted based on task complexity and input into the target large-scale language model. Finally, evaluation metrics and the robustly enhanced target large-scale language model are output. This invention supports multiple mainstream programming languages ​​and both open-source and closed-source models, exhibiting good versatility and scalability.

The invention discloses an effective code measurement method based on an AST abstract syntax tree. The effective code measurement method comprises the following steps that S1, source code files before and after change in a code warehouse are read; s2, tree transformation is carried out on the obtained source code file through ASTs, and tree ASTs before and after change are generated respectively; s3, traversing the tree-shaped AST before the change and the tree-shaped AST after the change, and generating a Tree diff change node through comparison; and S4, performing weighted calculation according to the generated Tree diff change node type to obtain an effective code quantity. Through AST-level comparison, semantic changes, such as code reconstruction (such as method movement) and function modification, which are difficult to detect by a traditional diff tool can be accurately identified, noise, such as spaces and dead codes, of a source code level can be eliminated, a compilation optimization technology can be applied, and evaluation is more accurate.

Techniques and solutions are provided for enhancing software security through autonomous adaptive code evolution. Code variants are generated and analyzed using various methods, such as static code analysis or execution in controlled environments, against known and predictive future threats or vulnerabilities to determine whether they exhibit improved security. Variants can be generated using techniques such as genetic programming, instruction substitution, control flow alteration, or dead code insertion. Types of code modifications that result in improved security are prioritized when generating variants. In one example, reinforcement learning is used to identify code adaptations that enhance security, including those which do so without overly compromising functionality or performance. Continuous performance monitoring can be used to help ensure that security adaptations do not degrade software functionality, and an intelligent rollback mechanism can be used to revert to a previous state if negative impacts are detected.