Date: Thu, 22 Mar 2012 05:06:33 +1100 (EST) From: Bruce Evans <brde@optusnet.com.au> To: Marius Strobl <marius@alchemy.franken.de> Cc: svn-src-head@FreeBSD.org, svn-src-all@FreeBSD.org, src-committers@FreeBSD.org Subject: Re: svn commit: r233273 - head/sys/conf Message-ID: <20120322045351.O2789@besplex.bde.org> In-Reply-To: <20120321165645.GE37560@alchemy.franken.de> References: <201203210850.q2L8olCj062739@svn.freebsd.org> <20120322031550.O2453@besplex.bde.org> <20120321165645.GE37560@alchemy.franken.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 21 Mar 2012, Marius Strobl wrote: > On Thu, Mar 22, 2012 at 03:28:04AM +1100, Bruce Evans wrote: >> On Wed, 21 Mar 2012, Marius Strobl wrote: >> >>> Log: >>> Exclude devices which are mutually exclusive with ATA_CAM. For better >>> or worse, the former are still built as modules as part of the LINT >>> builds >>> >>> Reviewed by: mav >>> MFC after: 1 week >> >> This breaks NOTES on these devices. The purpose of NOTES is to test as >> much as possible, not to provide notes. It is also supposed to test >> unusual cases. Thus if these unusualy devices actually conflict at >> compile time, then it is the usual ones that confict that should not >> be build by NOTES, since the usual ones get adequate testing in GENERIC. > > As I wrote above, the unusual case (!ATA_CAM) is still built as modules > as part of the LINT builds. R233271 actually was the minimal change to > still build both the usual (ATA_CAM) and the unusual case as apart of the > LINT builds. Not on systems which build the "correct" number of modules (that is, zero) as part of building kernels. Though LINT should arguably ignore the system's preferences for modules (as expressed by NO_MODULES in /etc/make.conf), it actually respects them, and it must not ignore the user's preferences for modules (as expressed by NO_MODULES in the environment). Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120322045351.O2789>