From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1LPEtB-0007bT-2n for qemu-devel@nongnu.org; Tue, 20 Jan 2009 06:35:25 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1LPEt8-0007aA-T1 for qemu-devel@nongnu.org; Tue, 20 Jan 2009 06:35:24 -0500 Received: from [199.232.76.173] (port=36513 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1LPEt7-0007Zy-Hj for qemu-devel@nongnu.org; Tue, 20 Jan 2009 06:35:21 -0500 Received: from gecko.sbs.de ([194.138.37.40]:22174) by monty-python.gnu.org with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1LPEt6-00046h-Va for qemu-devel@nongnu.org; Tue, 20 Jan 2009 06:35:21 -0500 Message-ID: <4975B6FB.6080302@siemens.com> Date: Tue, 20 Jan 2009 12:35:23 +0100 From: Jan Kiszka MIME-Version: 1.0 References: <1232448066-32209-1-git-send-email-amit.shah@redhat.com> In-Reply-To: <1232448066-32209-1-git-send-email-amit.shah@redhat.com> Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: [PATCH 1/2] Print asynchronous notifications on request Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Amit Shah , aliguori@us.ibm.com Amit Shah wrote: > This patch adds the ability to selectively enable asynchronous notifications > from individual qemu subsystems by a new 'notify' monitor command. > > This is helpful for programs currently parsing monitor output. > > A sample invocation will look like this: > > (qemu) > > (qemu) notify vnc on > > (qemu) # VNC: Closing down connection 127.0.0.1:1 > > Notice that the output is prefixed by '#'. Also, it will appear on the > line that has '(qemu) ' already output on it. I understand the need, but the result looks a bit ugly, at least to humans forced to parse it. If you get this while in the middle of type a command... Moreover, is it impossible that such an async notification is issued while some other subsystem is already dumping a multi-line message to the monitor (using multiple prints)? That would be really problematic. But if this approach is merged, then I would really say we need separately configurable monitor terminals, and notify should then only affect the issuing one. However, my feeling is that a real machine-dedicated and easily processable channel would be better suited for this use case. Jan -- Siemens AG, Corporate Technology, CT SE 2 Corporate Competence Center Embedded Linux