背景 在 FPGA 或者 ASIC 中，通常都需要使用 RAM，通过读口、写口或者读写口来进行访问。常见的配置有单读写口（1RW），一读一写（1R1W）等等，读口通常有 1 个周期的延时。那么，如果在同一个周期内，读口和写口访问了同一个地址，会发生什么呢？可能会想到几种情况：
读和写都失败，读出的数据未定义，数据没写进去 数据写进去了，读出的数据未定义 数据写进去了，读出了写之前的旧数据 数据写进去了，读出了同一个周期写入的新数据 下面以具体的例子来看看，实际情况是什么样子。
Xilinx FPGA 首先测试的是…

When optimizing software, we routinely measure the time that takes a given function or task. The typical assumption is that we get a normal distribution, and so we should therefore report the average time. I believe that this hidden assumption (normality)…

Generative artificial intelligence, and in particular ChatGPT, has taken the world by storm. Some intellectuals are proposing we create a worldwide ban on advanced AI research, using military operations if needed. Many leading researchers and pundits are…

前言作为IP vendor，ARM公司的产品受众广泛；除了为人所熟知的终端厂商之外，还有各式嵌入式用途。因此在设计核心时ARM需要兼顾各种使用场景，做到表现尽可能均衡的同时还需严控成本，ARM对微架构efficiency的执著…

拖延症拖到Gracemont都发布了，不能再拖了！ 阅读本文前，建议先看上一篇文章： XZiar：分支预测，uOP，乱序执行 Tremont是Intel在19年下半年公布的新一代atom微架构，上一代则是Goldmont/Goldmont Plus。 实际产…

Note the publish day of this article. There will be a proper one on Terascale 3 later on, don’t worry. But for now, happy April Fools Day! Prestige is everything for computer hardware manufacturers. A few percentage points may not make or break someone’s…

Open source FPGA-based NIC and platform for in-network compute - corundum/corundum

Optimizing compilers seek try to push as much of the computation as possible at compile time. In modern C++, you can declare a function as ‘constexpr’, meaning that you state explicitly that the function may be executed at compile time. The constexpr qualifier…

BUGGY and not compilable

modification:
- Add up down counter to TageEntry
- Rewrite update logic
- Remove legacy debug code

幕间在上篇中我们主要探究了鲲鹏920 TSV110微架构的取指前端和后端执行单元配置，下篇我们继续探究Mid Core、访存子系统、核外系统。 JamesAslan：华为鲲鹏920 TSV110微架构评测（上）:初露锋芒，砥砺前行Mid Core…

Personal knowledge wiki

Contribute to name99-org/AArch64-Explore development by creating an account on GitHub.