From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58175) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VdLuz-0007EY-1S for qemu-devel@nongnu.org; Mon, 04 Nov 2013 10:18:22 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VdLut-0000R6-3A for qemu-devel@nongnu.org; Mon, 04 Nov 2013 10:18:16 -0500 Received: from mx1.redhat.com ([209.132.183.28]:47680) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VdLus-0000Qw-Qt for qemu-devel@nongnu.org; Mon, 04 Nov 2013 10:18:11 -0500 Message-ID: <1383578284.8610.51.camel@nilsson.home.kraxel.org> From: Gerd Hoffmann Date: Mon, 04 Nov 2013 16:18:04 +0100 In-Reply-To: <20131104153515.0a5b9c32@nial.usersys.redhat.com> References: <1383051455-28188-1-git-send-email-imammedo@redhat.com> <20131029151047.GA20242@redhat.com> <20131029162825.6fe5a0ae@nial.usersys.redhat.com> <20131029185242.GC20848@redhat.com> <20131030142454.667f354b@nial.usersys.redhat.com> <20131030134838.GC2759@redhat.com> <1383143615.22559.42.camel@nilsson.home.kraxel.org> <20131030163858.3ebe0263@nial.usersys.redhat.com> <1383569283.8610.36.camel@nilsson.home.kraxel.org> <20131104153515.0a5b9c32@nial.usersys.redhat.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 0/2 v2] pc: inform SeaBIOS where 64-bit PCI hole begins List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Igor Mammedov Cc: pbonzini@redhat.com, "Michael S. Tsirkin" , afaerber@suse.de, aliguori@amazon.com, qemu-devel@nongnu.org On Mo, 2013-11-04 at 15:35 +0100, Igor Mammedov wrote: > On Mon, 04 Nov 2013 13:48:03 +0100 > Gerd Hoffmann wrote: > > > Hi, > > > > > > So maybe design that with memory hotplug in mind? Such as adding a new > > > > qemu-specific type QEMU_RAM_HOTPLUG? Which seabios could use to reserve > > > > the memory (but not add it to the e820 table for the guest)? > > > It will do job too. But extending semantics of standard table would be > > > confusing. Yes, Seabios will filter it out but it doesn't make table > > > less confusing. > > > > Was just thinking that it might be easier that way if we need e820 > > entries for hotplug memory address space _anyway_. > I don't think that we need e820 entries for hotplug memory reserved space as > e820 should. In case present at boot hotpluggable DIMMs would be needed in E820, > we can add them as usual E820_RAM entries. > > But regardless of what we do here it might be good keep option of adding non > standard entries in future, by filtering out unknown types in SeaBIOS. > > > > > > I'd prefer having a dedicated interface for it as a more clean solution. > > > > Agree. > > So back to naming question, would you agree to renaming fw_cfg to the last > Michael's suggestion "reserved-memory-end"? Fine with me. cheers, Gerd