From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hamish Moffatt Date: Thu, 2 Apr 2009 11:56:13 +1100 Subject: [Buildroot] svn commit: trunk/buildroot/toolchain: binutils external-toolchain gcc uClibc In-Reply-To: References: <20090130162827.2A05D765E9@busybox.osuosl.org> <20090330153259.0a7ee398@surf> <87ab73azsj.fsf@macbook.be.48ers.dk> <20090330175842.1960fed2@surf> <87prfyald1.fsf@macbook.be.48ers.dk> <20090401012841.585dc38e@surf> Message-ID: <20090402005613.GE29488@cloud.net.au> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On Tue, Mar 31, 2009 at 11:43:09PM -0300, Thiago A. Corr?a wrote: > Hi, > > 2009/3/31 Thomas Petazzoni : > > Le Mon, 30 Mar 2009 16:56:54 -0300, > > Thiago A. Corr?a a ?crit : > > > >> Would that be a bad thing? Even if it's not integrated but a script, > >> say: ./script/fix-external-toolchain.sh > >> > >> Before/after running kconfig. > >> > >> It beats having the user to guess or remember. > > > > Or maybe we could just *not* show these options, and let a script > > auto-detect the values. For uClibc, they can easily be guessed from > > some uClibc-config.h script available in the toolchain. > > It must be "guessed" before entering kconfig because of the package > options that depend on them. Specially large file support and IPv6. Should those options (particularly large file support) just become standard ie required? Hamish -- Hamish Moffatt VK3SB