From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Sat, 21 Sep 2013 05:41:45 +0200 Subject: [Buildroot] [autobuild.buildroot.net] Build results for 2013-09-19 In-Reply-To: <20130920063008.7C55952C101@lolut.humanoidz.org> References: <20130920063008.7C55952C101@lolut.humanoidz.org> Message-ID: <20130921054145.52b70d5e@skate> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello, On Fri, 20 Sep 2013 08:30:08 +0200 (CEST), Thomas Petazzoni wrote: > arc | host-gcc-initial-4.8-arc | NOK | http://autobuild.buildroot.net/results/ff903ce970ea652cc1aa57a5eb57e923dc4c5405/ > arc | host-gcc-initial-4.8-arc | NOK | http://autobuild.buildroot.net/results/d3b2e4e32d531c08c19d39b248898fe462b81d8f/ > arc | host-gcc-initial-4.8-arc | NOK | http://autobuild.buildroot.net/results/76a2ab46ef7183f14eec944db4a598325e00b503/ Hum, apparently on ARC, building libgcc requires having the C library headers. So, due to the change to build gcc in two passes instead of three, we're building libgcc in the first pass, before the C library is built, causing (I believe) the above failure. Mischa, if you have any insights about this, I'm interested :) Considering this problem, and the other SSP problems, I'm starting to wonder if my patch that changes the toolchain build mechanism to use two passes instead of three shouldn't be reverted for now, until I find the time to investigate all the problems that it is causing. Thomas -- Thomas Petazzoni, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com