Date: Mon, 8 Jun 1998 13:40:19 +0200 (CEST) From: Søren Schmidt <sos@FreeBSD.ORG> To: jkh@time.cdrom.com (Jordan K. Hubbard) Cc: sos@FreeBSD.ORG, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-sys@FreeBSD.ORG Subject: Re: cvs commit: src/sys/i386/isa/snd ad1848.c sb_dsp.c sound.c Message-ID: <199806081140.NAA03454@sos.freebsd.dk> In-Reply-To: <21391.897304637@time.cdrom.com> from "Jordan K. Hubbard" at "Jun 8, 98 04:17:17 am"
next in thread | previous in thread | raw e-mail | index | archive | help
In reply to Jordan K. Hubbard who wrote:
>
> 1. What we consider a "reasonable test" of any kernel change has always
> been a relatively clean build of LINT.
Resonable, but not enough for this kind of broad spread changes.
> 2. Doug built LINT.
So?
> 3. The sound code you had problems with is enabled in the LINT configuration
> (the old, not new Luigi code) and thus, assuming that everyone did their
> bit correctly, *was* tested by Doug and the failure must therefore be in
> the testing method itself.
You havn't been watching your commit mail then, there has been at least
a handfull (if not two full hands) of broken drivers that wasn't checked
for the ioctl change, the snd stuff was just a one off:
the devfs stuff was broken
the ELF activator was broken
the sound system was broken
the wd driver was broken
the wst driver was broken
DDB was broken.
Need I continue ??
> 4. If building LINT is no longer a sufficient test, we should determine
> what *is* a sufficient test before we start flaming people.
It has NEVER been sufficient to build LINT :), and it probably never will
be, there are too much interference between different drivers etc for
that to work. Find & grep is some of your helpers here, and so would a large
regressiontest framework be, but we don't have that. So one should at
least grep for the matter (here a simple grep for ioctl would have
done nicely), or have the code run through a couple of others as they
will most often have different setups and will probably spot some
of the offending bugs.
As usual we all make mistakes (and I've done my fair share), but generic
changes like these should be tested & reviewed by at least a couple
of others with different setups.
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team
Even more code to hack -- will it ever end
..
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199806081140.NAA03454>