From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Korsgaard Date: Tue, 16 Dec 2014 13:41:33 +0100 Subject: [Buildroot] [git commit] erlang: use libatomic_ops if available In-Reply-To: <20141216132618.5c19cb84@free-electrons.com> (Thomas Petazzoni's message of "Tue, 16 Dec 2014 13:26:18 +0100") References: <20141215215003.A758F80403@busybox.osuosl.org> <20141216132618.5c19cb84@free-electrons.com> Message-ID: <8761dbkbr6.fsf@dell.be.48ers.dk> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net >>>>> "Thomas" == Thomas Petazzoni writes: > Hello, > On Mon, 15 Dec 2014 22:36:24 +0100, Peter Korsgaard wrote: >> Erlang can either provide its own atomic_ops implementation or use the >> one in libatomic_ops. This patch ensures that it only uses the one from >> libatomic_ops if it is available. > Is this really what we want to do? In such cases, we normally always > force to use the "system" version of the library rather than the > built-in one. Why should Erlang be treated differently, especially if > using the "system" version of the library works fine? So in other words, make erlang select libatomic_ops? That sounds fine to me. -- Bye, Peter Korsgaard