From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:38755) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fYU4I-00044r-Nx for qemu-devel@nongnu.org; Thu, 28 Jun 2018 06:22:28 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fYU4H-0007n8-Ft for qemu-devel@nongnu.org; Thu, 28 Jun 2018 06:22:26 -0400 Date: Thu, 28 Jun 2018 11:22:16 +0100 From: Daniel =?utf-8?B?UC4gQmVycmFuZ8Op?= Message-ID: <20180628102216.GC3513@redhat.com> Reply-To: Daniel =?utf-8?B?UC4gQmVycmFuZ8Op?= References: <1530094388-607-1-git-send-email-hongbo.zhang@linaro.org> <20180627123116.GB2424@work-vm> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20180627123116.GB2424@work-vm> Subject: Re: [Qemu-devel] [PATCH] hw/arm: Add SBSA machine type List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Dr. David Alan Gilbert" Cc: Hongbo Zhang , peter.maydell@linaro.org, qemu-arm@nongnu.org, qemu-devel@nongnu.org On Wed, Jun 27, 2018 at 01:31:17PM +0100, Dr. David Alan Gilbert wrote: > * Hongbo Zhang (hongbo.zhang@linaro.org) wrote: > > This patch introduces a new Arm machine type 'SBSA' with features: > > - Based on legacy 'virt' machine type. The 'virt' machine type is absolutely *not* legacy. It is the preferred modern, legacy-free machine type when running virtual machines. Since aarch64 is greenfield, there is no compelling reason to emulate a specific physical machine for VMs, hence the 'virt' machine type. > > - Newly designed memory map. > > - EL2 and EL3 are enabled by default. > > - AHCI controller attached to system bus, and then CDROM and hard disc > > can be added to it. > > - EHCI controller attached to system bus, with USB mouse and key board > > installed by default. > > - E1000 ethernet card on PCIE bus. > > - VGA display adaptor on PCIE bus. > > - Default CPU type cortex-a57, 4 cores, and 1G bytes memory. > > - No virtio functions enabled, since this is to emulate real hardware. > > I'm a bit confused; do you have real modern ARM hardware that has an > e1000 on it? If I understand correctly, e1000 is the old PCI version > and the e1000e is at least the more modern PCIe version which makes > more sense if you're building on PCIe. Indeed, it makes little sense to default to e1000 if the goal is to make it legacy-free > However, if you: > a) Don't have real hardware with the e1000 on > b) Do have PCIe > > then to my mind it makes sense to use virtio-net-pci rather than > an e1000e. If it does down the virtio-* route though, I fail to see any point in adding the new machine type at all - 'virt' machine type is intended to be used for guests where everything is paravirtualized, ignoring physical hardware constraints. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|