ARC-101 · Processor Architecture · Foundation
CPU Pipelines & Microarchitecture — full syllabus
How a modern out-of-order core fetches, schedules and retires instructions, and why that determines the performance ceiling of your code.
Who this course is for
Kernel, driver and performance engineers who need to reason about CPU behaviour from the pipeline up rather than from folklore.
Prerequisites
- C programming fundamentals
- Basic Linux command line
- No prior microarchitecture background required
Course outline
Day 1 — The modern pipeline, fetch to retire
- Fetch, decode, rename, issue, execute, retire — the modern pipeline stage by stage
- Superscalar width and what actually limits it
- Reorder buffers and reservation stations: out-of-order execution with in-order retirement
- Following one instruction through a real core diagram
- ISA versus microarchitecture: what the manual promises and what the silicon does
Day 2 — Speculation and its costs
- Branch prediction: direction and target prediction, BTB, return stack
- The misprediction penalty in cycles, measured
- Data and control hazards and how the pipeline hides them
- Speculation side effects visible from software
- Reading perf stat output (IPC, branch-misses, stalls) against a pipeline model
Day 3 — Instruction-level parallelism in real code
- Dependency chains and the critical path
- Execution ports, port pressure and where ILP runs out
- What the compiler can and cannot do for the pipeline
- Comparing -O0/-O2 assembly against measured IPC
- Reading vendor optimisation manuals and microarchitecture whitepapers without getting lost
Hands-on labs
- Lab: measure IPC and front-end stalls with perf stat on tight loops and explain the numbers against a pipeline diagram
- Lab: construct a branch-misprediction microbenchmark and measure the penalty in cycles with perf stat -e branches,branch-misses
- Lab: build a pointer-chasing dependency chain, vary the chain length, and expose the load-to-use latency it reveals
- Lab: compare compiler output at -O0 and -O2 for a hot loop and correlate the instruction mix with measured IPC
- Lab: take one claim from a vendor optimisation manual and test it with a targeted microbenchmark
Capstone project
Build a cycle-accounted narrative for a small compute kernel: write the benchmark, state a pipeline-level hypothesis for where its cycles go, test it with PMU counters (cycles, instructions, branch-misses, stall events), and finish with an annotated assembly listing plus a counter table that either confirms the hypothesis or rejects it with evidence.
What you leave with
- A working pipeline model from fetch to retirement
- Hands-on fluency with perf stat and raw PMU event names
- Microbenchmarks that isolate branch, dependency and port effects
- A method for reading microarchitecture whitepapers critically
Upcoming dates
| Dates | Where | Seats | Early bird | Regular | |
|---|---|---|---|---|---|
| 18 Oct – 20 Oct 20263 full days | RiyadhIn person · KAFD Conference Centre | 10 of 14 | — | SAR 6,750 | |
| 25 Oct – 27 Oct 20263 full days | Kuwait CityIn person · Al Hamra Tower | 5 of 14 | — | KWD 560 | |
| 1 Nov – 3 Nov 20263 full days | MuscatIn person · Knowledge Oasis Muscat | 10 of 14 | — | OMR 690 | |
| 1 Nov – 8 Nov 20266 half-days | Gulf bandLive online · 09:00–13:00 GMT+3 | 10 of 20 | — | US$1,300 | |
| 2 Nov – 4 Nov 20263 full days | OttawaIn person · Kanata North Tech Park | 5 of 14 | — | CAD 2,450 | |
| 9 Nov – 11 Nov 20263 full days | TorontoIn person · MaRS Discovery District | 10 of 14 | CAD 2,200until 10 Oct | ||
| 9 Nov – 16 Nov 20266 half-days | Europe bandLive online · 09:00–13:00 CET | 15 of 20 | US$1,170until 10 Oct | ||
| 9 Nov – 16 Nov 20266 half-days | Americas bandLive online · 13:00–17:00 ET | 4 of 20 | US$1,170until 10 Oct | ||
| 16 Nov – 18 Nov 20263 full days | LondonIn person · Shoreditch Works | 5 of 14 | GBP 1,260until 17 Oct | ||
| 16 Nov – 18 Nov 20263 full days | BerlinIn person · Factory Görlitzer Park | 10 of 14 | EUR 1,490until 17 Oct |
Book a seat, or bring this course to your team
Seats can be reserved online; private delivery runs on-site or live online, adapted to your stack.
Questions about fit or prerequisites? Email hello@kernelsystems.academy. To save this syllabus, print this page to PDF from your browser.