From owner-freebsd-hackers Sat Jul 21 19:56: 9 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from sharmas.dhs.org (cpe-66-1-147-119.ca.sprintbbd.net [66.1.147.119]) by hub.freebsd.org (Postfix) with ESMTP id 8A9C037B405 for ; Sat, 21 Jul 2001 19:56:06 -0700 (PDT) (envelope-from adsharma@sharmas.dhs.org) Received: by sharmas.dhs.org (Postfix, from userid 500) id 508985DD97; Sat, 21 Jul 2001 19:56:20 -0700 (PDT) Date: Sat, 21 Jul 2001 19:56:20 -0700 From: Arun Sharma To: hackers@freebsd.org Subject: Re: libc_r, signals and modifying sigcontext Message-ID: <20010721195620.A32616@sharmas.dhs.org> References: <20010721191747.A32529@sharmas.dhs.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.15i In-Reply-To: <20010721191747.A32529@sharmas.dhs.org>; from arun@sharmas.dhs.org on Sat, Jul 21, 2001 at 07:17:47PM -0700 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Sat, Jul 21, 2001 at 07:17:47PM -0700, Arun Sharma wrote: > Greetings. I'm trying to port an application to FreeBSD. I have > a signal handler registered using signal(2). It modifies the > data pointed to by the third argument - of type sigcontext (specifically > sc_eip) - so that the execution would resume at a different point). > > However, when execution resumes, it resumes at the same point where > it was interrupted. A quick search of the archives brought up this > thread: Another data point: the problem doesn't happen with IBM's MxN pthread library ported to FreeBSD (mainly because it uses libc and not libc_r). http://oss.software.ibm.com/developerworks/opensource/pthreads/ I wonder how the scheduler activations based stuff is going to handle this case. -Arun To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message