Subject: Re: VERIFIED_EXEC to be removed from the config files (Re: CVS commit: src/sys/arch/i386/conf)
To: Elad Efrat <elad@NetBSD.org>
From: David Maxwell <firstname.lastname@example.org>
Date: 07/27/2006 10:25:29
> Bernd Ernesti wrote:
> > I don't think that all port master read source-changes, so they watch
> > all commits for other ports.
> > IMHO you should remove that entry because you removed that code
> > and added the options in the first place to these config files.
> huh? no code was removed, it was just done differently. the option can
> stay in the config, it just has no effect. now the veriexec code is
> compiled in only if you have the pseudo-device, which is enough.
> i'm not aware of a list that all port masters watch, but right now
> i think source-changes and current-users is enough coverage.
I think Bernd's point is that the change is a simple and mechanical one,
which doesn't require the intervention of the port masters.
In the same way that documentation should be kept up to date as changes
are made to the source code, any config file in-tree should be up to
date and specified correctly, corresponding with the current code.
Sending mail to a group of distracted individual volunteers guarantees
that some ports' config files will not be updated in a timely fashion -
or, perhaps at all - until someone notices it again later, probably by
wasting time being confused about how the code uses the supplied count...
It is appropriate for someone changing the code to also commit the updates
to the configs that reference that code at the same time. If NetBSD someday
switches to a changeset capable version control system, it should probably
become mandatory to do so, in order to maintain consistency.
David Maxwell, email@example.comfirstname.lastname@example.org -->
If you don't spend energy getting what you want,
You'll have to spend it dealing with what you get.