Re: [MAINTAINERS SUMMIT] The place of AI code review in the Linux Kernel process
Mark Brown <[email protected]> Thu, 23 Jul 2026 18:33:40 +0100
| Newsgroups | dev.linux.lists.ksummit |
|---|---|
| Message-ID | <[email protected]> |
--UF0bDZ5JprLErxBE Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Thu, Jul 23, 2026 at 01:08:42PM -0400, Rodrigo Vivi wrote: > I agree with Guenter approach here and in a matter of fact, I had just used > a few minutes ago. I had to review a series where I noticed Sashiko had a > couple of true finds. So, I just asked the developer to look to that report > and fix that before I waste my time with reviews that the tool could already > spot. > But I was already aware-of and okay-with this entire code design idea and > approach and this review would be more about the correctness. Yes, I think that's a very different submitter experience to something where there's no human intervention at all and much more reasonable. --UF0bDZ5JprLErxBE Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQEzBAABCgAdFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAmpiUHMACgkQJNaLcl1U h9BGPwf7Bh1esE3RdotSkWF3hvOVn5T2wy0OZCR0aGMEZ8ALageGC6rMMxsNAf+e G5mkz0W1GfXcP7cSua8BKJix85igBWWrubQOwnDG6KWxZfkhDJ/j4xY20XNlySfS WdHg69lDivbaGj3jT8yELzQaqlg2B3qfueol0Q6+kPL0o+ND1S17L13uyfKtG8qv O3wbe/FXCRZcrBwOU0UFswgntcSB/DIuN5tCundF7EQOLhxVf4xIwYex+QO/vkgS ODXEqaJrqMD6D60Ke/S+zrk/kLphi8BEkKN3B/vUj3xBWyFgB7GzwpZcXOR7CG4I y0BG9eC+wK7OWDwUQJfB0+cLty75MQ== =IL7R -----END PGP SIGNATURE----- --UF0bDZ5JprLErxBE--