On Sat, 2026-01-31 at 21:15 +0500, Islombek Ismoilov via Gcc wrote:
> Hello,

Hi Islombek

> 
> My name is Islombek, I am a nuclear physics and technology student
> with a
> strong interest in systems programming and compiler development, and
> I am
> interested in applying to GSoC 2026 with GCC. 

Excellent; welcome!

> I would like to ask for
> general advice regarding the project “Extend the static analysis
> pass”.

I would be the mentor of such a project (I wrote and maintain the
static analysis pass).

Note that there are 4 different ideas for projects on 
  https://gcc.gnu.org/wiki/SummerOfCode#Selected_Project_Ideas
relating to the static analyzer.   FWIW, of those, I'm most interested
in the format-string one (but am open to other ideas, if a candidate
has a particular thing they want to focus on and can convince me).

> 
> At this stage, I am reading the documentation and studying the
> existing
> implementation of the static analyzer, including its overall
> architecture
> and related code, in order to better understand the project.

Sounds like a good start.

> 
> I would appreciate any general advice or pointers from the community
> regarding this project.

I would recommend building GCC from source, and doing the other
exercises in the guide for new GCC contributors here:
https://gcc-newbies-guide.readthedocs.io/en/latest/
such as adding a "hello world"-style warning.

Once you've built GCC, try stepping through -fanalyzer in the debugger
on a really simple example (say, a few lines of C code with a trivial
bug in it), to see what it does with the user's code and how that turns
into a warning.

Good things to demonstrate in an application: 
* evidence of the ability to write and debug C++, 
* evidence of the ability to work with git (e.g. link to a repository,
if you have any code that you can link to).
* knowledge of compiler internals, if you already have any of that
* knowledge of static analysis theory, if you've studied that

Hope this is helpful
Dave

Reply via email to