https://jasper-aa64.github.io/zh/tags/x86/Recent content in X86 on JasperHugozh-CNTue, 15 Sep 2026 00:00:00 +0000https://jasper-aa64.github.io/zh/posts/memory-ordering-false-sharing-dependency-chains/Tue, 15 Sep 2026 00:00:00 +0000https://jasper-aa64.github.io/zh/posts/memory-ordering-false-sharing-dependency-chains/同一套缓存硬件带出两个不相干的问题。正确性：写先进 store buffer，别的核要过一段时间才看得见，靠 release/acquire 配对补上顺序。性能：就算每个值都读对了，两个核抢同一条缓存行（伪共享）照样慢，靠缓存行布局解决。