C++ Count Your Instructions Lab 1 Python lscpu machine instructions objdump perf terminal

Lab 1 · Count Your Instructions — Exploring Your Machine¶

This lab is different from the others: there is no breadboard and no motor. Today you explore what the computer is actually doing when it runs code. You will use the terminal to inspect your Raspberry Pi's hardware, look inside a compiled program, and count how many machine instructions two different programs execute to do the same job.

Where are the lab files?

The HOTIS-count-till-py.py, HOTIS-count-till-cpp.cc, and compiled files are pre-installed on your Pi image. Ask your instructor if you are not sure which directory to find them in.

Step 1 · Access the Terminal¶

Look for the grey icon with a white >_ in the top-left corner of the desktop and click it.

→ See Using the terminal if you need help opening or navigating it.

Step 2 · Check your CPU¶

lscpu

This shows details about the processor in your Pi. What architecture is it? How many cores does it have?

→ See Exploring your system for what these fields mean.

Step 3 · Check Memory¶

cat /proc/meminfo

Look for MemTotal. How much RAM does this machine have?

Step 4 · Check Storage¶

df -h

Find the row for /. How much total storage does the Pi have, and how much is used?

Step 5 · Show All Files¶

ls -al

This lists every file and folder in your current directory. You will use this constantly throughout all the labs.

Step 6 · Navigate the File System¶

Practice moving around:

cd <a directory name>

Move into a folder you saw in ls -al. To come back:

cd ..

.. always means "up one level." → See Using the terminal for more detail.

Step 7 · Look at the Programs¶

Navigate to the directory where the Lab 1 files live (ask your instructor if needed), then open the source files:

geany HOTIS-count-till-py.py
geany HOTIS-count-till-cpp.cc

Both programs count up to a number you give them. One is Python; one is C++. Read them side by side. How similar are they? How different?

Now look inside the compiled C++ file:

objdump -d HOTIS-count-till-cpp.o

This disassembles the compiled object file — it shows you the raw machine instructions the compiler produced. This is what the CPU actually runs. → See Compiled vs interpreted to understand what you are looking at.

Step 8 · Run the Programs¶

python HOTIS-count-till-py.py <a number of your choice>
./HOTIS-count-till-cpp.out <a number of your choice>

Replace <a number of your choice> with an actual integer, like 500 or 1000.

Try it — make a prediction

Before running Step 9, write down your guess: how many machine instructions do you think each program executes to count to that number? More for Python or more for C++? By how much?

Step 9 · Count the Machine Instructions¶

perf stat -e instructions:u python HOTIS-count-till-py.py <your number>
perf stat -e instructions:u ./HOTIS-count-till-cpp.out <your number>

perf stat -e instructions:u counts the actual number of user-space machine instructions that ran while executing the command.

Who guessed closest?
Are you surprised by how different (or similar) the two counts are?
What does this tell you about the difference between a compiled language and an interpreted one?

→ See Compiled vs interpreted for the explanation.

Step 10 · Shut Down¶

At the end of your session, shut down safely:

shutdown now

Never unplug the Pi while it's running

See Shutting down safely for why this matters.