From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bernhard Reutner-Fischer Date: Fri, 9 Oct 2009 12:55:00 +0200 Subject: [Buildroot] [PATCH 13/14] enable config.cache per default In-Reply-To: <20091009120035.0382fa34@surf> References: <53e08690da5a307328d4d5dc7d53865215ced09f.1254945750.git.rep.dot.nop@gmail.com> <730af4752b8da4a9b2a2b1583987899fa3eef334.1254945750.git.rep.dot.nop@gmail.com> <78b8984e033470578a90d58e533aa97c31af1071.1254945750.git.rep.dot.nop@gmail.com> <871vlekjie.fsf@macbook.be.48ers.dk> <20091009120035.0382fa34@surf> Message-ID: <20091009105500.GK2266@mx.loc> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On Fri, Oct 09, 2009 at 12:00:35PM +0200, Thomas Petazzoni wrote: >Le Wed, 07 Oct 2009 22:59:37 +0200, >Peter Korsgaard a ?crit : > >> Bernhard> It speeds up configuration considerably. >> Bernhard> If a package breaks due to wrong cache-entries, then the >> Bernhard> broken other package has to be fixed, not the innocent >> Bernhard> bystander! >> >> Bernhard> Put short: There is absolutely _no_ need to ever turn >> Bernhard> the cache off unless you hack on autotools itself and goof. >> >> In that case, why do we even have the config option? > >Originally, because when the change was introduced, some packages were >failing due to incorrect dependencies. And the contributor didn't want >to break many packages at once and preferred to have a intermediate >step where this is an option. > >I think we can now get rid of the option. Or introduce some kind of >?Buildroot Hacking? menu/submenu where we put this kind of very >advanced options that normal users shouldn't use, but that can be >helpful for debugging. And when doing support, we could also ask users >to enable the option that disables the cache to see if it fixes a >reported failure. i'd just leave it as knob, default y. Note: http://git.uclibc.org/~aldot/git/buildroot/commit/?h=misc&id=f4221a72ee2d10aa2376be1aae4ead05a44802de