The core of the NPU compilation process is the “operator”. An operator can be understood as each computational step in the model, such as convolution, pooling, and fully connected layers. The compiler determines the destination of each operator through a multi-stage decision process.The core process of separating tasks between the CPU and NPU can be … Read more
Hello everyone, I am a programmer who loves to share. I am happy to share my experiences and insights from my work. -begin- Day 1: The -mcpu Option of Cross Toolchains – “Tailoring” the Instruction Set for Target CPUs In embedded development, advanced options of cross toolchains (such as arm-linux-gnueabihf-gcc) can make the compiled programs … Read more
[Problem Description] Little A has an array consisting of n non-negative integersa=[a₁，a₂，a₃，…，aₙ] . He will repeatedly perform the following operations on the array a until the array a only contains 0. In one operation, Little A will sequentially complete the following three steps: 1、Find the largest integer in the array a, and denote its index … Read more
Introduction Starting with a painful downtime incident. The response time of the online business system skyrocketed from 200ms to 8 seconds, yet the CPU usage (65%), memory (30% free), and disk I/O appeared “normal”. The core issue: judging system health based on a single metric ignores the complexity of Linux performance, which is a complex … Read more
In embedded development, we often need to parse serial data, such as reading the return information from ESP8266, identifying module status, and waiting for <span>"OK"</span> or <span>"ERROR"</span> to appear. However, this returned data often does not arrive all at once, but rather—in segments fed into the queue. Thus, the concept of window search comes into … Read more
1. Essence of Macros The preprocessor is the first step in the C language compilation process, performing text replacement on the source code before actual compilation. Macro definitions are a form of string replacement mechanism that does not involve type checking or semantic analysis. Compilation Process: Source Code → Preprocessing → Compilation → Assembly → … Read more
Faker has won six championships!It’s quite emotional; I used to watch Faker win championships back in school, and he is still dominating the arena! The trophies he holds are not just a testament to his victories but also the best proof of the esports spirit — remaining humble at the peak and resilient in the … Read more
Efficient programming techniques in C include: Bit Manipulation: Prefer using bit manipulation instead of multiplication and division. Global Variables: Accessing global variables is usually faster than accessing local variables. Pointers: Accessing array elements using pointers is generally faster than using array indices. Algorithm Optimization: Arrange branches and loops logically to reduce unnecessary calculations. Macros: Use … Read more
In the world of game development, C++ is undoubtedly the king of programming languages. From Cyberpunk 2077 to Fortnite, almost all mainstream game engines (such as Unreal Engine and the underlying Unity) rely on the powerful performance and flexibility of C++. So, how does C++ “ignite” game engines? Where are the secrets of high-performance rendering … Read more
Optimization of File Descriptors and Connection Counts in Linux In Linux system administration, file descriptors (File Descriptor) and connection counts are key parameters for performance optimization. A file descriptor is a handle that a process uses to access resources, while the connection count directly affects the capability of high-concurrency applications. According to a report by … Read more