Re: git: 225639e7db68 - main - vt: Disable bell by default

From: Warner Losh <imp_at_bsdimp.com>
Date: Mon, 01 Nov 2021 02:40:31 UTC
On Sun, Oct 31, 2021 at 8:27 PM Kevin P. Neal <kpn@neutralgood.org> wrote:

> On Fri, Oct 22, 2021 at 10:55:40PM +0200, Emmanuel Vadot wrote:
> >  If I knew that this change would have caused that much mails I would
> > have send a mail to current or whatever before and left the review
> > opened for more time.
>
> Is this exposing an issue with the review process? I don't know the FreeBSD
> review process, but I do know the LLVM one. Now, LLVM has a different set
> of people in the project. That means that what works over there might not
> be a good fit here. I thought I'd mention it just the same.
>
>
> On the LLVM project they don't allow code that is up for review to be
> pushed without an explicit signoff from one of the reviewers. Period.
>
> There's _no_ putting code up for review, waiting, and then going ahead and
> pushing.
>
> If there's no feedback for a while the person asking for the review will
> add a message every week or two that says just "ping". This often nudges
> people to get to the review. Not always, and it can take a while to get
> code through, but that's the process.
>
> Changes are allowed without a review if they are "obvious". If someone
> pushes a change that another committer doesn't feel is obvious then the
> change is reverted, usually by the person who cares. Personally, I've never
> had this happen, but my non-review pushes are usually either new tests or
> fixes to broken builds.
>

I clicked approve on this quickly after manu posted it. There was little
other comment.
Seemed like a boring change.

A better review process wouldn't have helped. Dozens of other boring
commits have
a very similar profile.

Warner