Memory Ordering in Modern Microprocessors

Paul E. McKenney · 2007

Linux R ○ has supported a large number of SMP systems based on a variety of CPUs since the 2.0 kernel. Linux has done an excellent job of abstracting away differences among these CPUs, even in kernel code. One important difference is how CPUs allow memory accesses to be reordered in SMP systems. SMMP Hardware Memory accesses are among the slowest of a CPU’s operations, due to the fact that Moore’s law has increased CPU instruction performance at a much greater rate than it has increased memory performance. This difference in performance increase means that memory operations have been getting increasingly expensive compared to simple register-to-register instructions. Modern CPUs sport increasingly large caches in order to reduce the

Read the paper · More papers on PaperTik