Skip to content

Latest commit

 

History

History
173 lines (129 loc) · 4.77 KB

README.md

File metadata and controls

173 lines (129 loc) · 4.77 KB

CFiddle: A Tool For Studying Small Compiled Programs

CircleCI Binder Documentation Status

CFiddle is a tool for studying the compilation, execution, and performance of smallish programs written in compiled languages like C, C++, or Go. If you want to know what the compiler does to your code, how your code interacts with hardware, and why it's slow, CFiddle can help.

CFiddle makes it easy to ask and answer interesting questions about what happens to programs as they go from source code to running program. CFiddle can run on its own, but it is built to work with Jupyter Notebook/Jupyter Lab to support interactive exploration. It provides first-class support for accessing hardware performance counters.

It's features include:

  1. Support for compiled languages like C, C++, and Go.
  2. Multiple architecture (x86, ARM, PowerPC, etc.) and compiler (gcc and clang) support.
  3. Easy access to OS and hardware performance counters.
  4. Control Flow Graph (CFG) generation from compiled code.
  5. Easy support for varying build-time and run-time paremeters.
  6. Easy, unified parameter and data gathering across building and running code.
  7. Works great with Pandas and Jupyter Notebook/Lab.

Trying CFiddle

The best way to learn about CFiddle is to try it. There are several ways to do this:

  1. Run the examples on Binder.org (this can take a while to load, and performance counters don't work.).
  2. You visit https://try-cfiddle.nvsl.io (to try CFiddle on x86). Performance counters are fully functionaly.
  3. Run it locally with Docker (--privileged is required to access the performance counters):
docker run -it --privileged --publish published=8889,target=8888 stevenjswanson/cfiddle:latest jupyter lab --LabApp.token='sesame'

and then visit http://localhost:8889/lab/tree/README.ipynb -- the token is "sesame".

Documentation

Here's the documentation.

Examples

What Does a for loop look like in assembly?

>>> from cfiddle import * 
>>> sample = code(r""" 
...    extern "C"
...    int loop() {
...    int sum = 0;
...	   for(int i = 0; i < 10; i++) {
... 		sum += i;
...    }
...	   return sum;
... }
... """)
>>> asm = build(sample)[0].asm("loop")
>>> print(asm) # doctest: +SKIP
loop:
.LFB0:
    .cfi_startproc
    endbr64
    pushq    %rbp
    .cfi_def_cfa_offset 16
    .cfi_offset 6, -16
    movq    %rsp, %rbp
    .cfi_def_cfa_register 6
    movl    $0, -8(%rbp)
    movl    $0, -4(%rbp)
.L3:
    cmpl    $9, -4(%rbp)
    jg    .L2
    movl    -4(%rbp), %eax
    addl    %eax, -8(%rbp)
    addl    $1, -4(%rbp)
    jmp    .L3
.L2:
    movl    -8(%rbp), %eax
    popq    %rbp
    .cfi_def_cfa 7, 8
    ret
    .cfi_endproc

Or, if you prefer a CFG:

>>> build(sample)[0].cfg("loop", "readme_loop.png") 
'readme_loop.png'

CFG Example

What Does -O3 Do To That Loop?

>>> asm = build(sample, build_parameters=arg_map(OPTIMIZE="-O3"))[0].asm("loop")
>>> print(asm) # doctest: +SKIP
loop:
.LFB0:
    .cfi_startproc
	endbr64
	movl	$45, %eax
	ret
	.cfi_endproc
	

Local Installation

CFiddle depends on some system packages and python's wheel. Setup a virtual environment:

python -m venv cfiddle-venv

So you can install the system packages CFiddle needs. Check install_prereqs.sh to see what this includes. It uses apt-get.

sudo bash
. cfiddle-venv/bin/activate	
# export CFIDDLE_INSTALL_CROSS_COMPILERS=yes # uncomment if you want all the cross compilers
./install_prereqs.sh
exit

Install cfiddle:

. cfiddle-venv/bin/activate
pip install .

Run the tests:

make test

Common Problems

CFiddle needs LD_LIBRARY_PATH set properly to work, and it can't set it itself reliably. If you get

OSError: libcfiddle.so: cannot open shared object file: No such file or directory

You can update LD_LIBRARY_PATH with:

$(set-cfiddle-ld-path)

Saving Your Work

If you want to save changes you make to any of the examples, you'll need to run docker something like this (this assumes you're in the root of your cfiddle directory):

docker run -it -d --publish published=8888,target=8888 --mount type=bind,source=$PWD,dst=/home/jovyan/cfiddle -w /home/jovyan/  stevenjswanson/cfiddle:devel  jupyter lab --LabApp.token=''