Re: [MAINTAINERS SUMMIT] Other LLM-related topics - tags, newcomers, etc

--M//rVEmCGvuCPPZK
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Wed, Jul 22, 2026 at 10:20:55AM -0400, Trond Myklebust wrote:
> On Tue, 2026-07-21 at 17:57 -0700, Linus Torvalds wrote:

> > And the corollary to that basic issue of trust is that invasive
> > contributions from unknown people should be much less likely to be
> > accepted.

> > (Here I define "invasive contribution" as something that is hard to
> > verify as a patch - there can be large and invasive patches that are
> > easy to verify with tooling, so it's not necessarily about the size
> > per se).

> In my experience, "large and invasive" hasn't been the biggest problem.
> It's rather the fact that the people submitting have no understanding
> of the subsystems they are patching, and so they (and the AI) overstate
> the claims of importance. Suddenly, I'm seeing a slew of "security
> vulnerabilities" that are Cced to the stable kernel channel, and
> sometimes the security channels, and therefore scream "Drop everything
> you're doing and look at me now!".

> IOW: the problem is intrusiveness onto the maintainer's schedule, which
> also eventually leads to fatigue. Particularly so, when none of those
> claims actually stand up to scrutiny. The patch itself may be nominally
> correct, but it's never a "drop everything you're doing" issue.

Or just simple volume of non-repetitive but generally perfectly valid
changes.  It's not even that things are invasive, it's just that there's
a lot of them and you don't get the benefit of pattern matching that
tends to happen with determinalistic tooling.

--M//rVEmCGvuCPPZK
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEzBAABCgAdFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAmpiIewACgkQJNaLcl1U
h9DJ5gf/c+8AwIEhYYuQ7o0jdMzhwzM4HG7MKUudpNLRGVa7iDk9x13BPkZ8s6ub
+DLC03VmkcyrUgrfIOobnA2h/lsN0oxgG//ujVsbXT0AcNMDBDzLHVO78BhJnsbJ
Hpu9JD9d8yFj87Bpm3EsZ51mDR8tGIe9AXFbBIxmCeTyJSDVR70B7Cmz7Oe9gAei
+1uhaIeoCFOwcWcMcfTE+RePexxquQmT4jKb4sCudlgeJEWmjvX9K2nEEKXP7B5E
V3bfKXsC+VUhZ5YDuqqKCeYlXUaLKifMxRLJGEtduscoDKXVTYbDbNDbsFCofYNJ
KnCquEYLTr03zfOM2csLeau0TZHzug==
=R7d8
-----END PGP SIGNATURE-----

--M//rVEmCGvuCPPZK--