From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48964) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1etW4u-00054a-8u for qemu-devel@nongnu.org; Wed, 07 Mar 2018 05:13:45 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1etW4p-0006L4-AE for qemu-devel@nongnu.org; Wed, 07 Mar 2018 05:13:44 -0500 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:43148 helo=mx1.redhat.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1etW4p-0006Kn-5N for qemu-devel@nongnu.org; Wed, 07 Mar 2018 05:13:39 -0500 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.rdu2.redhat.com [10.11.54.5]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id C95C776FB9 for ; Wed, 7 Mar 2018 10:13:38 +0000 (UTC) Date: Wed, 7 Mar 2018 18:13:25 +0800 From: Peter Xu Message-ID: <20180307101325.GP17720@xz-mi> References: <20180216131625.9639-1-dgilbert@redhat.com> <20180216131625.9639-21-dgilbert@redhat.com> <20180302075107.GM27381@xz-mi> <20180305195513.GW3131@work-vm> <20180306033736.GA3615@xz-mi> <20180306105418.GD3096@work-vm> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20180306105418.GD3096@work-vm> Subject: Re: [Qemu-devel] [PATCH v3 20/29] postcopy: postcopy_notify_shared_wake List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Dr. David Alan Gilbert" Cc: qemu-devel@nongnu.org, maxime.coquelin@redhat.com, marcandre.lureau@redhat.com, imammedo@redhat.com, mst@redhat.com, quintela@redhat.com, aarcange@redhat.com On Tue, Mar 06, 2018 at 10:54:18AM +0000, Dr. David Alan Gilbert wrote: [...] > > Basically above was what I thought - to record the faulted addresses > > with specific PostcopyFD when page fault happened, then we may know > > which page(s) will a PostcopyFD need. But when with that, we'll > > possibly need a lock to protect the information (or any other sync > > method). > > OK, but I think you're suggesting building a whole new data structure to > know which ones need notifying; that sounds like a lot of extra > complexity for not much gain. Yes we may need a new structure (or just a list of addresses?), and indeed I have no idea on how that would help us. I think it depends on how many useless wakeup we will have, and how expensive is each of such a wakeup notification. Again, I think current solution is good enough as long as we don't see explicit blocker on performance side, and we can rethink that when really needed. Thanks, -- Peter Xu