From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:57507) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZMB1A-0006oK-Um for qemu-devel@nongnu.org; Mon, 03 Aug 2015 04:22:45 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZMB13-0003V9-PB for qemu-devel@nongnu.org; Mon, 03 Aug 2015 04:22:44 -0400 Received: from mx1.redhat.com ([209.132.183.28]:58465) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZMB13-0003V4-Jo for qemu-devel@nongnu.org; Mon, 03 Aug 2015 04:22:37 -0400 Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23]) by mx1.redhat.com (Postfix) with ESMTPS id 2B94419F265 for ; Mon, 3 Aug 2015 08:22:37 +0000 (UTC) Date: Mon, 3 Aug 2015 16:22:34 +0800 From: Fam Zheng Message-ID: <20150803082234.GA30561@ad.nay.redhat.com> References: <20150731174542.44862e3a@markmb_rh> <20150803030906.GA13938@ad.nay.redhat.com> <20150803095238.663a7bee@markmb_rh> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20150803095238.663a7bee@markmb_rh> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] Modularizing QEMU RFC List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Marc =?iso-8859-1?Q?Mar=ED?= Cc: qemu-devel On Mon, 08/03 09:52, Marc Mar=ED wrote: > So any other ideas to reduce the library overhead are appreciated. It would be interesting to see your profiling on the library loading over= head. For example, how much does it help to reduce the library size, and how mu= ch does it help to reduce the # of libraries? The protocol drivers are modularized for the sake of library dependencies= , so they should stay that way. However, we can "sensibly" combine all non-nat= ive format drivers (VMDK, VHDX, ...) into a cold-formats.so (if it turns out = that loading one big .so is much faster than loading separate ones). But we s= hould leave qcow2 as a separate one for obvious reasons, or make a hot-formats.= so with one or two other formats if that makes more sense. With that, for the first step, we can lazy load the cold-formats.so whene= ver we need to probe or a non-qcow2 format is involved. Then on top of that we c= an implement what Peter has suggested. Fam