Subject: Re: veriexec by default (Re: CVS commit: src/sys/arch)
To: Elad Efrat <elad@NetBSD.org>
From: Simon Burge <simonb@NetBSD.org>
List: tech-kern
Date: 11/27/2006 10:19:57
Elad Efrat wrote:

> YAMAMOTO Takashi wrote:
> >> YAMAMOTO Takashi wrote:
> >>>>>> i object.  please revert.
> >>>>> I'm gonna have to let other people voice in before I revert this one.
> >>>> ok.
> >>> i still think it should be reverted.
> >> http://mail-index.netbsd.org/source-changes/2006/10/27/0043.html
> > 
> > do you mean it shouldn't be reverted because garbled followed you?
> 
> close: I mean it shouldn't be reverted because it doesn't seem removing
> it reached a consensus, but quite the opposite. I voiced my opinion,
> you voiced yours, we decided to let others voice theirs, that's what
> happened.

FWIW, I've got

	no options      FILEASSOC       # fileassoc(9) - required for Veriexec
	no pseudo-device veriexec

in my kernel config, which just includes GENERIC then tweaks a few
different things.  I didn't figure the (admittedly unmeasured) hit on
exec times was worth it for a feature I never plan on using.

Simon.