Date: Wed, 11 Jul 2007 23:46:51 -0700 From: Brian Somers <brians@ca.sophos.com> To: mjacob@FreeBSD.org Cc: cvs-src@FreeBSD.org, src-committers@FreeBSD.org, Brian Somers <brian@FreeBSD.org>, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/net if.c if_var.h src/sys/netinet in.c in_var.h Message-ID: <20070711234651.5efc1c66@dev.lan.Awfulhak.org> In-Reply-To: <20070706190843.X78793@ns1.feral.com> References: <200707070054.l670slr1007427@repoman.freebsd.org> <20070706183517.512b6dab@conflict> <20070706190843.X78793@ns1.feral.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 6 Jul 2007 19:08:57 -0700 (PDT) mjacob@FreeBSD.org wrote:
>
> Cool. Can we put this in regression tests?
To be honest, I've never looked in tools/regression...
until now. It looks like a bit of a mess.
According to the README (last updated in 2001)
it should be possible to ``prove -r'' everything.
This is far from true, so forgive me for ranting
if I've got the wrong end of the stick.
Here's what I'd expect:
- ``prove -r'' works
- ``make test'' works in a similar way to ``prove -r'',
sets PATH to /usr/obj/... and fails if tests fail.
- .t files build test programs as required.
- .t files skip() tests (or plan a different number)
when required resources aren't available.
- comments on ok/not ok lines are mandatory
It would also be nice to bring libtap into the base system.
If there are no objections, I'll look at doing this after
-current un-slushes. If there are any suggestions about
other improvements, please follow-up to freebsd-arch.
--
Brian Somers Tel: +1 604 484 6434
Mob: +1 604 315 1343
Sophos - security and control Web: www.sophos.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070711234651.5efc1c66>