From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Sat, 20 Aug 2016 22:27:20 +0200 Subject: [Buildroot] [PATCH] pinentry: fix build issue with gcc 5.x In-Reply-To: References: <1471701636-30275-1-git-send-email-thomas.petazzoni@free-electrons.com> Message-ID: <20160820222720.51b932bf@free-electrons.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello, On Sat, 20 Aug 2016 13:19:02 -0700, Khem Raj wrote: > > +# pinentry uses some std::string functionality that needs C++11 > > +# support when gcc >= 5.x. This should be removed when bumping > > +# pinentry, since newer versions no longer use std::string. > > +ifeq ($(BR2_TOOLCHAIN_GCC_AT_LEAST_5),y) > > +PINENTRY_CONF_ENV = CXXFLAGS="$(TARGET_CXXFLAGS) -std=gnu++11" > > +endif > > Should this package be restricted to gcc5 and newer only, I assume > it will break with gcc 4.x too and c++11 may not be there, gcc5 atleast > has the support its just not default We have lots of gcc 4.x toolchains, and we haven't had any build failure. I believe it builds fine with gcc 4.x. The code requiring C++11 that causes the build failure with gcc 5.x is in libstdc++ itself. As I discussed with Yann on IRC, I don't really understand why some code in libstdc++ uses C++11, without gcc 5.x defaulting to C++11. Maybe you have a clue? Thanks, Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com