Date: Wed, 21 Nov 2012 18:59:37 -0700 From: Warner Losh <imp@bsdimp.com> To: Benjamin Kaduk <kaduk@MIT.EDU> Cc: svn-doc-head@freebsd.org, svn-doc-all@freebsd.org, doc-committers@freebsd.org, Eitan Adler <eadler@freebsd.org> Subject: Re: svn commit: r40117 - in head/en_US.ISO8859-1/books/developers-handbook: kernelbuild policies tools Message-ID: <CE9A6172-0450-4F60-9234-F8B159260A7C@bsdimp.com> In-Reply-To: <alpine.GSO.1.10.1211212031500.2164@multics.mit.edu> References: <201211211357.qALDvDsP064264@svn.freebsd.org> <alpine.GSO.1.10.1211211255140.2164@multics.mit.edu> <CAF6rxgmjGS3ahcrVsgzPqo%2B-ep9PyOmPn6oZBEH0AWJYDChzCw@mail.gmail.com> <alpine.GSO.1.10.1211212031500.2164@multics.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Nov 21, 2012, at 6:33 PM, Benjamin Kaduk wrote: > On Wed, 21 Nov 2012, Eitan Adler wrote: >=20 >> On 21 November 2012 13:14, Benjamin Kaduk <kaduk@mit.edu> wrote: >>> This seems to remove the last documentation of manually invoking = config(8) >>> to build a kernel, which seems worthy of explicit mention, and = perhaps >>> broader discussion. >>=20 >> This should probably be documented in config(1) or some other "how = the > (config(8)) >> build process works" document. So far as I could tell, the old = method >> should be considered an implementation detail, not an alternative. >=20 > The old method does not require building a toolchain or buildenv, if I = remember correctly. No, it does not. Nor does it require anything more than is on the = system right now. Often, this is sufficient. Sometimes it isn't. The = buildkernel version was done to keep people from shooting themselves... Warner >> That said I agree this probably deserves some broader consideration, >> so perhaps doc@ is the best place to discuss? Or would a developer >> focussed ML like hackers@ be better? >=20 > I think hackers would be better. >=20 > -Ben
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CE9A6172-0450-4F60-9234-F8B159260A7C>