Date: Sun, 16 Nov 2003 18:54:32 -0700 (MST) From: "M. Warner Losh" <imp@bsdimp.com> To: jroberson@chesapeake.net Cc: wollman@khavrinen.lcs.mit.edu Subject: Re: cvs commit: src/sys/conf kern.post.mk kmod.mk Message-ID: <20031116.185432.132432651.imp@bsdimp.com> In-Reply-To: <20031116191819.O10222-100000@mail.chesapeake.net> References: <200311170009.hAH09n7u073008@khavrinen.lcs.mit.edu> <20031116191819.O10222-100000@mail.chesapeake.net>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <20031116191819.O10222-100000@mail.chesapeake.net>
Jeff Roberson <jroberson@chesapeake.net> writes:
: > <<On Mon, 17 Nov 2003 11:04:20 +1100 (EST), Bruce Evans <bde@zeta.org.au> said:
: >
: > > Agreed, this is required the support the bug.
: >
: > Not everyone agrees that this is a bug.
: >
: > I would, however, be willing to stipulate that it is worth limiting
: > the number of option combinations a third-party vendor would need to
: > support.
:
: Anymore than 1 per release is too much test load for an officially
: supported product. We're only going to further deter companies from
: providing binary only drivers if they aren't going to work with most
: kernels. Furthermore, we're likely to crash if the driver is not
: compatible with the kernel.
Agreed. It was only with great reluctance that I enabled it. This
will at least give us experience before pulling the trigger on a full
blown kernel module build system re-write.
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20031116.185432.132432651.imp>