From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41012) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gAeoL-0001FX-Qj for qemu-devel@nongnu.org; Thu, 11 Oct 2018 13:31:46 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gAeoL-0000TZ-2h for qemu-devel@nongnu.org; Thu, 11 Oct 2018 13:31:45 -0400 Received: from mail-oi1-x244.google.com ([2607:f8b0:4864:20::244]:34565) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1gAeoK-0000Sk-ML for qemu-devel@nongnu.org; Thu, 11 Oct 2018 13:31:44 -0400 Received: by mail-oi1-x244.google.com with SMTP id v69-v6so7725793oif.1 for ; Thu, 11 Oct 2018 10:31:43 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: References: <20181009232607.15521-1-crosa@redhat.com> <20181009232607.15521-5-crosa@redhat.com> <835fcb24-5832-9e6c-9f57-0364f0bf849a@redhat.com> <95a8561c-0511-6a8a-a3eb-25d556029e7a@redhat.com> <20181010134656.GE5738@habkost.net> <47f72e98-335f-b3c9-54c7-4c8a468d85d6@redhat.com> <26fe2055-77c2-45fa-4e88-62101d70b88b@redhat.com> <20181010142840.GH5738@habkost.net> <118ed8f7-5947-12c6-f3f6-b10597f37aac@redhat.com> <00aa913d-a7e9-3c5d-34a8-fc9ff59091db@redhat.com> From: Peter Maydell Date: Thu, 11 Oct 2018 18:31:22 +0100 Message-ID: Content-Type: text/plain; charset="UTF-8" Subject: Re: [Qemu-devel] [PATCH v2 4/7] scripts/qemu.py: set predefined machine type based on arch List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Cleber Rosa Cc: Eduardo Habkost , Fam Zheng , =?UTF-8?Q?Philippe_Mathieu=2DDaud=C3=A9?= , QEMU Developers , Laszlo Ersek , =?UTF-8?Q?Philippe_Mathieu=2DDaud=C3=A9?= , Stefan Hajnoczi , Caio Carrara , =?UTF-8?B?QWxleCBCZW5uw6ll?= On 10 October 2018 at 20:54, Cleber Rosa wrote: > On 10/10/18 2:07 PM, Peter Maydell wrote: >> We should aim to test the machines we care about (regardless >> of what architectures they are), rather than thinking about it >> in terms of "testing architectures X, Y, Z", I think. >> > > To me it's clear that: > > 1) I lack a complete understanding of what "we care about" > 2) It's easier to start with something, and tweak it to taste Right. Effectively you're already defining a set of "machines we care about" -- it's just that rather than explicitly saying 'we care about x86-64 pc, i386 pc, aarch64 virt, ppc spapr, ...' you're implicitly doing it by defining a 'default machine' for various architectures and a list of architectures. I think you should just list the machines you're testing with, instead. I don't mind if we start with a list of what we're testing that's small (or even wrong!) -- we can expand and correct it later. I'd just like the list to be the correct shape. thanks -- PMM