Re: Core Toolchain Infrastructure - Services for glibc

[Konstantin Ryabitsev <konstantin@linuxfoundation.org>]

On Tue, Jul 18, 2023 at 02:26:44PM +0200, Jakub Jelinek wrote:
> > This would mean porting AdaCore hooks to a patchwork trybot.  This would be
> > an acceptable solution for glibc, but I'm not sure how useful this would be
> > on the whole since gcc doesn't use patchwork as extensively at the moment.
> > Also, we need to figure out who's going to do this.
> 
> It is definitely not acceptable for gcc, we strongly rely on server side
> pre-commit hooks for various different tasks.

I'm not saying "no hooks at all," but the general rule is:

- a hook can run checks on the contents of the commit message
- a hook can query external systems for validation results (with a reasonably
  short timeout)

The core principle is that hooks should be fast and shouldn't have a large
attack surface, because hooks are running with full access to the underlying
repository. Any CI builds or other validations should run prior to the push
and visible to the committer before the push is attempted. I mention patchwork
because it's a system we use, but other options exist (e.g. gerrit, though
I wouldn't use it just for CI purposes).

Regards,
-K