From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gustavo Zacarias Date: Sun, 30 Dec 2012 09:03:27 -0300 Subject: [Buildroot] [autobuild.buildroot.net] Build results for 2012-12-29 In-Reply-To: <87r4m7uacj.fsf@dell.be.48ers.dk> References: <20121230073456.2BCF452C08C@lolut.humanoidz.org> <87r4m7uacj.fsf@dell.be.48ers.dk> Message-ID: <50E02D8F.7020504@zacarias.com.ar> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On 12/30/2012 07:02 AM, Peter Korsgaard wrote: > These two looks related to your recent gnutls version bump. Care to have > a look? Running a failing config now, i don't think i'll hit it barebones though since they look like a pretty regular config. Probably have to debootstrap. > This is probably also related to it, but perhaps that gets fixed by > Thomas' e17 update. Seems unlikely from looking at the source of ecore. Since people have been using and accustomed to gnutls-2.x which required libgcrypt they've been including it blindly. I'll test to see if it really requires libgcrypt (like ntfs-3g) or it's just probably following old gnutls instructions and fix accordingly. > Did you have a look at the host-gawk issue? Yes, it seems peko-specific so my hunch says it's a distro toolchain thing. I've asked Thomas and it seems it's runnig gcc 4.3.2. Looks like the new gawk DFA is triggering a bug or using a newer feature that it shouldn't. Couldn't reproduce yet, so an old gentoo stage or debootstrap might help me. So also looking into it. A puny dual-core won't help me with getting these things solved quickly though, except libecore :) Regards.