From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1O5VQb-00050H-FO for qemu-devel@nongnu.org; Fri, 23 Apr 2010 22:49:09 -0400 Received: from [140.186.70.92] (port=59985 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1O5VQa-000509-EJ for qemu-devel@nongnu.org; Fri, 23 Apr 2010 22:49:09 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1O5VQZ-0007zr-9y for qemu-devel@nongnu.org; Fri, 23 Apr 2010 22:49:08 -0400 Received: from mail2.shareable.org ([80.68.89.115]:33961) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1O5VQZ-0007zj-4X for qemu-devel@nongnu.org; Fri, 23 Apr 2010 22:49:07 -0400 Date: Sat, 24 Apr 2010 03:49:02 +0100 From: Jamie Lokier Subject: Re: [Qemu-devel] Atomicity of i386 guest atomic instructions Message-ID: <20100424024902.GI15349@shareable.org> References: <4BD1D883.9080601@jermar.eu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexander Graf Cc: Jakub Jermar , qemu-devel@nongnu.org Alexander Graf wrote: > They should be atomic. TCG SMP swaps between different vCPUs only > after translation blocks are done. In fact, the only way I'm aware > of to stop the execution of a TB mid-way is a page fault. A page fault would interrupt it if the atomic is implemented as a read followed by a write, and the write faults. > You can as always check things with the -d parameter. -- Jamie