Received: from mail.pbcl.net ([88.198.119.4] helo=hetzner.pbcl.net) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1QKCzd-00087m-5B for openembedded-devel@lists.openembedded.org; Wed, 11 May 2011 19:14:37 +0200 Received: from cambridge.roku.com ([81.142.160.137] helo=[172.30.1.145]) by hetzner.pbcl.net with esmtpsa (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.69) (envelope-from ) id 1QKCYm-0000xR-IO for openembedded-devel@lists.openembedded.org; Wed, 11 May 2011 18:46:52 +0200 From: Phil Blundell To: openembedded-devel@lists.openembedded.org In-Reply-To: <1305062085.9982.62.camel@mattotaupa> References: <1305062085.9982.62.camel@mattotaupa> Date: Wed, 11 May 2011 18:11:51 +0100 Message-ID: <1305133911.2421.29.camel@phil-desktop> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Subject: Re: minimal-uclibc: systemd fails with `src/manager.c:257:42: error: 'EPOLL_CLOEXEC' undeclared (first use in this function)` X-BeenThere: openembedded-devel@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list Reply-To: openembedded-devel@lists.openembedded.org List-Id: Using the OpenEmbedded metadata to build Distributions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 11 May 2011 17:14:37 -0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit On Tue, 2011-05-10 at 23:14 +0200, Paul Menzel wrote: > I am no uClibc expert, but do you know if uClibc will provide > `EPOLL_CLOEXEC` in the near future? How should systemd fix that? It's not just EPOLL_CLOEXEC: uClibc seems to be missing epoll_create1() entirely. As far as I know there's nothing complex about that, though: it's just a standard syscall wrapper, so I don't know why it would have been deliberately left out. I guess you'd have to ask the uClibc h4x0rs about that. There probably is no trivial workaround available to systemd. If you wanted a quick hack to make it work then you could call syscall(SYS_epoll_create1, O_CLOEXEC) manually. Or you could perhaps eliminate the usage of EPOLL_CLOEXEC in favour of a separate fcntl() call, but that may not be safe if systemd is multithreaded. p.