From nobody Wed Nov 27 12:37:06 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1698403411; cv=none; d=zohomail.com; s=zohoarc; b=gn0WnW7cnheRzze7iENIZVoOVREDMEkB28x1ZC8xlT0d7UXd2hdkKljaLTnuQMjYdJ981eZcTAfQDFFl71HzstTEyzgYKMhx2eUOdMlLyWnKqkk6ooxEHoyueuPgrITXAXWHHJzMkcm4GRRp0yVlqfxy79v5kMc9f+/ZwNmLNnU= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1698403411; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=scdmRXEhmflRiIEVcQ1tB0CpGGSODn7Cz+FtlXseOA4=; b=jPyPaT4EMNRDMIEsj69WYY9rbq5GBVf1trw8sdY5J7pl2CvNBD1f46YMsC46w8lv0QTfDXDgdTz79pUTV4qP/LPeEEfk+qLI6GjHd0majqftTiwbGK4Xzpmep9+MDQ6rVPEswgBcCTuknKniDKzwp461Zupn96T5Qbo8RCEdNfA= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1698403411610281.0186237358687; Fri, 27 Oct 2023 03:43:31 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qwKIm-0008JS-Hi; Fri, 27 Oct 2023 06:42:52 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qwKIk-0008Ig-PQ for qemu-devel@nongnu.org; Fri, 27 Oct 2023 06:42:50 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qwKIh-00036D-U2 for qemu-devel@nongnu.org; Fri, 27 Oct 2023 06:42:50 -0400 Received: from mail-pg1-f198.google.com (mail-pg1-f198.google.com [209.85.215.198]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-369-8A1AaBeWMsmJLBvzPeYhZw-1; Fri, 27 Oct 2023 06:42:45 -0400 Received: by mail-pg1-f198.google.com with SMTP id 41be03b00d2f7-5b7dfda133dso1649675a12.0 for ; Fri, 27 Oct 2023 03:42:45 -0700 (PDT) Received: from localhost.localdomain ([115.96.30.106]) by smtp.googlemail.com with ESMTPSA id g4-20020a056a001a0400b0068fe7e07190sm1074273pfv.3.2023.10.27.03.42.40 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 27 Oct 2023 03:42:43 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1698403366; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding; bh=scdmRXEhmflRiIEVcQ1tB0CpGGSODn7Cz+FtlXseOA4=; b=DwoHvq3qhE6Ht+bn22OzW5dDSWcJJ6E0DnOnihJaKubKXirGjVQGirqlbROS4WjtxFrWCk ng0DQaDxSdP7RltJuJjZLUCB/08/qwWRK9Ns/6ShCJu3mgFnaMIIT/RqFBfX+gw8oNwAqh Cw0FFjwsNGE8ECQGe+H6aEC5UG0/Apc= X-MC-Unique: 8A1AaBeWMsmJLBvzPeYhZw-1 X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1698403364; x=1699008164; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=scdmRXEhmflRiIEVcQ1tB0CpGGSODn7Cz+FtlXseOA4=; b=Kx/o07bxCHulohWYmVee2YTspcz92OWhAEKmLC2tC1CGbVpN91kTENoBvLrKX3xgOo l/HOWFQDVLUaNOOK8CFIkInDPLiZvVpvkmINwicAjYRh5K0LFb+qmNh/oLkbAwTKyho2 BXaG+NXSvZWGdkeoOGYswdOo2jnVNznfitQxGgJHQHnArnK/d2sLGSdppQoFmSqt3WhR LE+4al7xb+z8D0KuZh94Mf+XUEG31RPnZhwbRdQa94KhvSxqWZmnQCFgnWQsE1kXN8WL l/ITjKeN3hQ91aDsh/FCbbAW64xxdVbncg/X80NTKTWLgdW9XA/a7OV5MKtIbhN3zMwF Z1yw== X-Gm-Message-State: AOJu0YwMh62+X0mCYk25WT9TdSBKEjgKg83juhKvSTHXG9r6as3t22G6 MPULk35Ik2xHiUm+V0Bf7O05SstsNhw0RJWN4H2XjNllwnKjGZSJFzkrIS2mbcdgRqW7JwadIK8 U2u4GnMfWZWta5h4= X-Received: by 2002:a05:6a20:9189:b0:156:e1ce:d4a1 with SMTP id v9-20020a056a20918900b00156e1ced4a1mr2933657pzd.9.1698403364036; Fri, 27 Oct 2023 03:42:44 -0700 (PDT) X-Google-Smtp-Source: AGHT+IFoerb+D0RJm2pKx2TX8tRHMkHwY58T3veBuprdOnplg+dutaeCFFCbOFvhGK/G1wnyLyxOeA== X-Received: by 2002:a05:6a20:9189:b0:156:e1ce:d4a1 with SMTP id v9-20020a056a20918900b00156e1ced4a1mr2933631pzd.9.1698403363496; Fri, 27 Oct 2023 03:42:43 -0700 (PDT) From: Ani Sinha To: Cleber Rosa , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wainer dos Santos Moschetta , Beraldo Leal Cc: Ani Sinha , imammedo@redhat.com, David Hildenbrand , qemu-devel@nongnu.org Subject: [PATCH v3] tests/avocado: add test to exercize processor address space memory bound checks Date: Fri, 27 Oct 2023 16:12:18 +0530 Message-ID: <20231027104219.36248-1-anisinha@redhat.com> X-Mailer: git-send-email 2.42.0 MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=anisinha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H4=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1698403414117100003 Content-Type: text/plain; charset="utf-8" QEMU has validations to make sure that a VM is not started with more memory (static and hotpluggable memory) than what the guest processor can address directly with its addressing bits. This change adds a test to make sure QEMU fails to start with a specific error message when an attempt is made to start a VM with more memory than what the processor can directly address. The test also checks for passing cases when the address space of the proces= sor is capable of addressing all memory. Boundary cases are tested. CC: imammedo@redhat.com CC: David Hildenbrand Signed-off-by: Ani Sinha --- tests/avocado/mem-addr-space-check.py | 331 ++++++++++++++++++++++++++ 1 file changed, 331 insertions(+) create mode 100644 tests/avocado/mem-addr-space-check.py Changelog: v3: added pae tests as well. v2: added 64-bit tests. Added cxl tests. Sample run: $ ./pyvenv/bin/avocado run tests/avocado/mem-addr-space-check.py --tap - 1..14 ok 1 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_ps= e36 ok 2 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_pae ok 3 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_pen= tium_pse36 ok 4 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_pen= tium_pae ok 5 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_pen= tium2 ok 6 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_no= npse36 ok 7 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_tc= g_q35_70_amd ok 8 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_tc= g_q35_71_amd ok 9 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_tcg= _q35_70_amd ok 10 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_tc= g_q35_71_amd ok 11 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_tc= g_q35_71_intel ok 12 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_t= cg_q35_71_amd_41bits ok 13 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_low_t= cg_q35_intel_cxl ok 14 tests/avocado/mem-addr-space-check.py:MemAddrCheck.test_phybits_ok_tc= g_q35_intel_cxl diff --git a/tests/avocado/mem-addr-space-check.py b/tests/avocado/mem-addr= -space-check.py new file mode 100644 index 0000000000..6ded11d658 --- /dev/null +++ b/tests/avocado/mem-addr-space-check.py @@ -0,0 +1,331 @@ +# Check for crash when using memory beyond the available guest processor +# address space. +# +# Copyright (c) 2023 Red Hat, Inc. +# +# Author: +# Ani Sinha +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +from avocado_qemu import QemuSystemTest +import signal + +class MemAddrCheck(QemuSystemTest): + # first, lets test some 32-bit processors. + # for all 32-bit cases, pci64_hole_size is 0. + def test_phybits_low_pse36(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + With pse36 feature ON, a processor has 36 bits of addressing. So i= t can + access up to a maximum of 64GiB of memory. Memory hotplug region b= egins + at 4 GiB boundary when "above_4g_mem_size" is 0 (this would be tru= e when + we have 0.5 GiB of VM memory, see pc_q35_init()). This means total + hotpluggable memory size is 60 GiB. Per slot, we reserve 1 GiB of = memory + for dimm alignment for all newer machines (see enforce_aligned_dimm + property for pc machines and pc_get_device_memory_range()). That l= eaves + total hotpluggable actual memory size of 59 GiB. If the VM is star= ted + with 0.5 GiB of memory, maxmem should be set to a maximum value of + 59.5 GiB to ensure that the processor can address all memory direc= tly. + Note that 64-bit pci hole size is 0 in this case. If maxmem is set= to + 59.6G, QEMU should fail to start with a message "phy-bits are too = low". + If maxmem is set to 59.5G with all other QEMU parameters identical= , QEMU + should start fine. + """ + self.vm.add_args('-S', '-machine', 'q35', '-m', + '512,slots=3D1,maxmem=3D59.6G', + '-cpu', 'pentium,pse36=3Don', '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + def test_phybits_low_pae(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + With pae feature ON, a processor has 36 bits of addressing. So it = can + access up to a maximum of 64GiB of memory. Rest is the same as the= case + with pse36 above. + """ + self.vm.add_args('-S', '-machine', 'q35', '-m', + '512,slots=3D1,maxmem=3D59.6G', + '-cpu', 'pentium,pae=3Don', '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + def test_phybits_ok_pentium_pse36(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Setting maxmem to 59.5G and making sure that QEMU can start with t= he + same options as the failing case above with pse36 cpu feature. + """ + self.vm.add_args('-machine', 'q35', '-m', + '512,slots=3D1,maxmem=3D59.5G', + '-cpu', 'pentium,pse36=3Don', '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") + + def test_phybits_ok_pentium_pae(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Test is same as above but now with pae CPUID turned on. Setting + maxmem to 59.5G and making sure that QEMU can start fine with the + same options as the case above. + """ + self.vm.add_args('-machine', 'q35', '-m', + '512,slots=3D1,maxmem=3D59.5G', + '-cpu', 'pentium,pae=3Don', '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") + + def test_phybits_ok_pentium2(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Pentium2 has 36 bits of addressing, so its same as pentium + with pse36 ON. + """ + self.vm.add_args('-machine', 'q35', '-m', + '512,slots=3D1,maxmem=3D59.5G', + '-cpu', 'pentium2', '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") + + def test_phybits_low_nonpse36(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Pentium processor has 32 bits of addressing without pse36 or pae + so it can access up to 4 GiB of memory. Setting maxmem to 4GiB + should make QEMU fail to start with "phys-bits too low" message. + """ + self.vm.add_args('-S', '-machine', 'q35', '-m', + '512,slots=3D1,maxmem=3D4G', + '-cpu', 'pentium', '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + # now lets test some 64-bit CPU cases. + def test_phybits_low_tcg_q35_70_amd(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + For q35 7.1 machines and above, "above_4G" memory starts at 1 TiB + boundary for AMD cpus (default). Lets test without that case. + For q35-7.0 machines, "above 4G" memory starts are 4G. + pci64_hole size is 32 GiB. Since TCG_PHYS_ADDR_BITS is defined to + be 40, TCG emulated CPUs have maximum of 1 TiB (1024 GiB) of + directly addressible memory. + Hence, maxmem value at most can be + 1024 GiB - 4 GiB - 1 GiB per slot for alignment - 32 GiB + 0.5 GiB + which is equal to 987.5 GiB. Setting the value to 988 GiB should + make QEMU fail with the error message. + """ + self.vm.add_args('-S', '-machine', 'pc-q35-7.0', '-m', + '512,slots=3D1,maxmem=3D988G', + '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + def test_phybits_low_tcg_q35_71_amd(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + AMD_HT_START is defined to be at 1012 GiB. So for q35 machines + version > 7.0 and AMD cpus, instead of 1024 GiB limit for 40 bit + processor address space, it has to be 1012 GiB , that is 12 GiB + less than the case above. + Make sure QEMU fails when maxmem size is 976 GiB (12 GiB less + than 988 GiB). + """ + self.vm.add_args('-S', '-machine', 'pc-q35-7.1', '-m', + '512,slots=3D1,maxmem=3D976G', + '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + def test_phybits_ok_tcg_q35_70_amd(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Same as q35-7.0 AMD case except that here we check that QEMU can + successfully start when maxmem is < 988G. + """ + self.vm.add_args('-S', '-machine', 'pc-q35-7.0', '-m', + '512,slots=3D1,maxmem=3D987.5G', + '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") + + def test_phybits_ok_tcg_q35_71_amd(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Same as q35-7.1 AMD case except that here we check that QEMU can + successfully start when maxmem is < 976G. + """ + self.vm.add_args('-S', '-machine', 'pc-q35-7.1', '-m', + '512,slots=3D1,maxmem=3D975.5G', + '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") + + def test_phybits_ok_tcg_q35_71_intel(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Same parameters as test_phybits_low_tcg_q35_71_amd() but use + Intel cpu instead. QEMU should start fine in this case as + "above_4G" memory starts at 4G. + """ + self.vm.add_args('-S', '-cpu', 'Skylake-Server', + '-machine', 'pc-q35-7.1', '-m', + '512,slots=3D1,maxmem=3D976G', + '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") + + def test_phybits_low_tcg_q35_71_amd_41bits(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + AMD processor with 41 bits. Max cpu hw address =3D 2 TiB. + "above_4G" memory starts at 1 TiB for q35-7.1 machines. So with + pci_64_hole size at 32 GiB, maxmem should be 991.5 GiB with 1 GiB = per + slot for alignment and 0.5 GiB as static memory for the VM + (1024 - 32 - 1 + 0.5). With 992 GiB, QEMU should fail to start. + """ + self.vm.add_args('-S', '-cpu', 'EPYC-v4,phys-bits=3D41', + '-machine', 'pc-q35-7.1', '-m', + '512,slots=3D1,maxmem=3D992G', + '-display', 'none', + '-object', 'memory-backend-ram,id=3Dmem1,size=3D1= G', + '-device', 'virtio-mem-pci,id=3Dvm0,memdev=3Dmem1= ') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + def test_phybits_low_tcg_q35_intel_cxl(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + cxl memory window starts after memory device range. Here, we use 1= GiB + of cxl window memory. 4G_mem end aligns at 4G. pci64_hole is 32 Gi= B and + starts after the cxl memory window. + So maxmem here should be at most 986 GiB considering all memory bo= undary + alignment constraints with 40 bits (1 TiB) of processor physical b= its. + """ + self.vm.add_args('-S', '-cpu', 'Skylake-Server,phys-bits=3D40', + '-machine', 'q35,cxl=3Don', '-m', + '512,slots=3D1,maxmem=3D987G', + '-display', 'none', + '-device', 'pxb-cxl,bus_nr=3D12,bus=3Dpcie.0,id= =3Dcxl.1', + '-M', 'cxl-fmw.0.targets.0=3Dcxl.1,cxl-fmw.0.size= =3D1G') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.wait() + self.assertEquals(self.vm.exitcode(), 1, "QEMU exit code should be= 1") + self.assertRegex(self.vm.get_log(), r'phys-bits too low') + + def test_phybits_ok_tcg_q35_intel_cxl(self): + """ + :avocado: tags=3Dmachine:q35 + :avocado: tags=3Darch:x86_64 + + Same as above but here we do not reserve any cxl memory window. He= nce, + with the exact same parameters as above, QEMU should start fine ev= en + with cxl enabled. + """ + self.vm.add_args('-S', '-cpu', 'Skylake-Server,phys-bits=3D40', + '-machine', 'q35,cxl=3Don', '-m', + '512,slots=3D1,maxmem=3D987G', + '-display', 'none', + '-device', 'pxb-cxl,bus_nr=3D12,bus=3Dpcie.0,id= =3Dcxl.1') + self.vm.set_qmp_monitor(enabled=3DFalse) + self.vm.launch() + self.vm.shutdown() + self.assertEquals(self.vm.exitcode(), -signal.SIGTERM, + "QEMU did not terminate gracefully upon SIGTERM") --=20 2.42.0