Computer Architecture
From first principles to the out-of-order engines in shipping processors.
66 chapters5 parts≈ 34.6 h read
A sixty-six chapter course that starts at what a transistor does when you put a voltage on its gate and ends inside the out-of-order engines of shipping processors. Concrete numerical example first, abstract definition second, connection to the application third.
Sixty-six chapters in five Parts, meant to be read in order, each assuming the one before it and nothing else. The depth ramps deliberately: the opening Parts assume a first-semester undergraduate who has never seen a logic gate, and the closing Parts read at research level. Each Part ends with case studies of real silicon and with project and lab chapters that turn the reading into something you build. The Hardware Interview covers the same ground from the other side. Register renaming here is a mechanism to derive, and there it is a question to answer at a whiteboard. Each chapter's Extended By section points to the interview chapters built on it, and the interview chapters point back. Everything rests on public sources: published architecture manuals, peer-reviewed papers, vendor disclosures at venues such as Hot Chips, and open-source RTL. Where a detail would be proprietary, the treatment turns to a published academic equivalent or a textbook formulation instead.
Table of Contents
5 parts · 66 chapters- 01The Computing Stackbeginner62 min read
- 02Trends, Constraints, and Quantitative Principlesbeginner50 min read
- 03Number Systems and Encodingsbeginner50 min read
- 04Boolean Algebra and Combinational Logicbeginner49 min read
- 05Digital Building Blocksbeginner58 min read
- 06Sequential Logic and Timingbeginner53 min read
- 07Integer Arithmetic Inside the CPUbeginner28 min read
- 08Floating-Point Arithmeticbeginner20 min read
- 09From Logic to Siliconbeginner29 min read
- 10Hardware Description Languagesbeginner26 min read
- 11Project: Building a 32-Bit ALU in Chiselbeginner24 min read
- 12Lab: The Open-Source HDL Workflowbeginner25 min read
- 01Anatomy of an Instruction Setintermediate61 min read
- 02RISC vs CISC and Other Classificationsintermediate53 min read
- 03RISC-V: A Modern Open ISAintermediate38 min read
- 04ARM A64: The Mobile and Server Workhorseintermediate42 min read
- 05x86-64: Intel and AMD's Shared Legacyintermediate33 min read
- 06Addressing Modes Side-by-Sideintermediate34 min read
- 07Procedure Calling Conventionsintermediate35 min read
- 08Privilege, Exceptions, and System Instructionsintermediate32 min read
- 09Vector and SIMD Programming Modelsintermediate38 min read
- 10Case Studies: The Public ISA Manualsintermediate35 min read
- 11Project: A Two-Pass Assembler for RV32IMCintermediate29 min read
- 12Lab: Toolchain Setup with Spike, QEMU, and Cross-Toolchainsintermediate33 min read
- 01The Single-Cycle Datapathintermediate31 min read
- 02The Control Unitintermediate29 min read
- 03The Multi-Cycle Datapathintermediate26 min read
- 04Introduction to Pipeliningintermediate30 min read
- 05The Classic 5-Stage Pipeline (IF-ID-EX-MEM-WB)intermediate32 min read
- 06Pipeline Hazardsintermediate29 min read
- 07Exception and Interrupt Handling in a Pipelined CPUintermediate28 min read
- 08Pipeline Optimization Beyond 5 Stagesintermediate21 min read
- 09Case Study: Open-Source In-Order Coresintermediate26 min read
- 10Project: Building a 5-Stage Pipelined RV32IM CPU in Chiselintermediate34 min read
- 11Lab: Verification and Cycle-Accurate Simulationintermediate26 min read
- 01The Memory Walladvanced29 min read
- 02Cache Fundamentalsadvanced32 min read
- 03Cache Performance Analysisadvanced26 min read
- 04Advanced Cache Topicsadvanced27 min read
- 05Prefetchingadvanced28 min read
- 06Virtual Memoryadvanced29 min read
- 07TLBs and Address Translationadvanced23 min read
- 08DRAM and Memory Controllersadvanced27 min read
- 09NVM, Storage, and Storage-Class Memoryadvanced24 min read
- 10I/O Architectureadvanced24 min read
- 11Case Study: Cache Hierarchies in Public Disclosuresadvanced30 min read
- 12Project: A Trace-Driven Cache Simulator in C++advanced29 min read
- 13Lab: ChampSim, DRAMsim3, and gem5 Memory Modelingadvanced23 min read
- 01Limits of In-Order Executionadvanced29 min read
- 02Tomasulo's Algorithmadvanced29 min read
- 03Register Renamingadvanced28 min read
- 04The Reorder Buffer (ROB)advanced25 min read
- 05Issue Queues and Schedulersadvanced26 min read
- 06Load-Store Queue Designadvanced28 min read
- 07Branch Prediction Foundationsadvanced25 min read
- 08Modern Branch Predictorsadvanced22 min read
- 09Frontend Bandwidthadvanced22 min read
- 10Vector Microarchitectureadvanced22 min read
- 11Simultaneous Multithreading (SMT)advanced34 min read
- 12Power, Frequency, and Thermaladvanced29 min read
- 13Case Study: BOOM (Berkeley Out-of-Order Machine)advanced26 min read
- 14Case Study: ARM Cortex-X3 and Neoverse N2advanced24 min read
- 15Case Study: Intel Golden Cove and Redwood Coveadvanced28 min read
- 16Case Study: AMD Zen 4advanced27 min read
- 17Project: A Branch Predictor Evaluatoradvanced28 min read
- 18Lab: gem5 Out-of-Order Modelingadvanced23 min read