From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49097) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WNth7-0004oS-Tx for qemu-devel@nongnu.org; Wed, 12 Mar 2014 20:40:24 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WNth6-0003cx-73 for qemu-devel@nongnu.org; Wed, 12 Mar 2014 20:40:21 -0400 Received: from b.painless.aa.net.uk ([2001:8b0:0:30::51bb:1e34]:50312) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WNth6-0003cO-0Y for qemu-devel@nongnu.org; Wed, 12 Mar 2014 20:40:20 -0400 Received: from 0.2.f.d.0.a.2.9.b.c.8.2.e.b.c.6.9.0.3.d.f.7.a.c.0.b.8.0.1.0.0.2.ip6.arpa ([2001:8b0:ca7f:d309:6cbe:28cb:92a0:df20] helo=athena.zubnet.me.uk) by b.painless.aa.net.uk with esmtps (TLSv1:AES128-SHA:128) (Exim 4.72) (envelope-from ) id 1WNth2-0000t6-6q for qemu-devel@nongnu.org; Thu, 13 Mar 2014 00:40:18 +0000 Received: from sdb by athena.zubnet.me.uk with local (Exim 4.82) (envelope-from ) id 1WNth1-0004ZS-Jm for qemu-devel@nongnu.org; Thu, 13 Mar 2014 00:40:15 +0000 Date: Thu, 13 Mar 2014 00:40:15 +0000 From: Stuart Brady Message-ID: <20140313004015.GC13298@zubnet.me.uk> References: <1394410549-13751-1-git-send-email-afaerber@suse.de> <1394410549-13751-20-git-send-email-afaerber@suse.de> <20140312230853.GB13298@zubnet.me.uk> <5320F4C5.2040307@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <5320F4C5.2040307@suse.de> Subject: Re: [Qemu-devel] [PATCH qom-cpu v2 19/40] cpu: Move breakpoints field from CPU_COMMON to CPUState List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org On Thu, Mar 13, 2014 at 12:59:01AM +0100, Andreas Färber wrote: > Am 13.03.2014 00:08, schrieb Stuart Brady: > > On Mon, Mar 10, 2014 at 01:15:28AM +0100, Andreas Färber wrote: > >> Most targets were using offsetof(CPUFooState, breakpoints) to determine > >> how much of CPUFooState to clear on reset. Use the next field after > >> CPU_COMMON instead, if any, or sizeof(CPUFooState) otherwise. > > > > Would it not be easier and more readable to add a field replacing > > 'breakpoints' just for this purpose, at least for the time being? > > > > I'm guessing CPU_COMMON_TLB will be getting cleaned up too at some point > > so is it really worth avoiding the tiny amount of bloat this imposes? > > Given the bad timing and having already done the work for v1, I am > reluctant to change the somewhat tested code unless there is a bug? No bug — and I've checked carefully. I was just concerned as it seemed a little fragile, but as the timing is bad (I now see that rc1 is due on the 19th) then that seems fair enough. -- Cheers, Stuart