From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Tue, 21 Nov 2017 23:52:56 +0100 Subject: [Buildroot] Analysis of build results for 2017-11-19 In-Reply-To: References: <20171120070004.B482020743@mail.free-electrons.com> <20171120233253.541c48ad@windsurf.lan> Message-ID: <20171121235256.75c6e007@windsurf.lan> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello, On Tue, 21 Nov 2017 23:24:13 +0100, Romain Naour wrote: > > On Mon, 20 Nov 2017 08:00:04 +0100 (CET), Thomas Petazzoni wrote: > > > >> sh4 | cppcms-1.0.5 | NOK | http://autobuild.buildroot.net/results/7d2cd95bc75250b47f5c9947cd996852925f6e0b | ORPH > > > > /tmp/ccFVFYgX.s: Assembler messages: > > /tmp/ccFVFYgX.s:34102: Error: pcrel too far > > > > Toolchain issue that nobody is ever going to fix I guess. > > > > This issue was reported a long time ago on binutils (for SuperH), and > > was apparently fixed. But now it appears again. > > > > What should we do? > > I don't really know about sh4 issue but a new cppcms 1.1.x version will be > available in near future [1]. > > The 1.0.x version is getting old. > > [1] https://sourceforge.net/projects/cppcms/files/cppcms We should also try with the latest binutils and gcc, and see if the problem still exists. If it does, we should report it upstream. > > This is a build issue caused by glibc 2.26. Pascal, since you added > > google-breakpad, could you have a look? Perhaps Romain Naour can help > > as well, since he bumped glibc to 2.26. > > As for gcc, we need to replace struct ucontext with ucontext_t. > I'll try to backport this patch [2] > > [2] > https://github.com/google/breakpad/commit/bddcc58860f522a0d4cbaa7e9d04058caee0db9d Great, thanks! Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com