From patchwork Wed Jan 4 12:39:39 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Andrew Cooper X-Patchwork-Id: 9496675 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 2BD0D606B5 for ; Wed, 4 Jan 2017 12:48:27 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 1D99B27F85 for ; Wed, 4 Jan 2017 12:48:27 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 1281827F8C; Wed, 4 Jan 2017 12:48:27 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-4.2 required=2.0 tests=BAYES_00, RCVD_IN_DNSWL_MED autolearn=ham version=3.3.1 Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) (using TLSv1.2 with cipher AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 9CEFB27F88 for ; Wed, 4 Jan 2017 12:48:26 +0000 (UTC) Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.84_2) (envelope-from ) id 1cOkxX-0005Zs-L8; Wed, 04 Jan 2017 12:46:27 +0000 Received: from mail6.bemta3.messagelabs.com ([195.245.230.39]) by lists.xenproject.org with esmtp (Exim 4.84_2) (envelope-from ) id 1cOkxW-0005Wu-MK for xen-devel@lists.xen.org; Wed, 04 Jan 2017 12:46:26 +0000 Received: from [85.158.137.68] by server-5.bemta-3.messagelabs.com id 59/3C-19998-1AEEC685; Wed, 04 Jan 2017 12:46:25 +0000 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFmpkkeJIrShJLcpLzFFi42JxWrohUnfhu5w Ig01nBSyWfFzM4sDocXT3b6YAxijWzLyk/IoE1oy/35exFZzRrmg9soCxgfG9fBcjJ4eEgL/E y/Xr2UFsNgF9id0vPjGB2CIC6hKnOy6ydjFycDAL+EkceuALYgoLBEjsnJMJYrIIqEgsfcgMU swr4Cnxd3U7E8RAOYnzx3+CxTmB4s82rmMEsYUEPCSmXFzGDmGrSVzrv8QO0SsocXLmExYQm1 lAQuLgixfMExh5ZyFJzUKSWsDItIpRvTi1qCy1SNdEL6koMz2jJDcxM0fX0MBYLze1uDgxPTU nMalYLzk/dxMjMGgYgGAHY+MXp0OMkhxMSqK8fe05EUJ8SfkplRmJxRnxRaU5qcWHGGU4OJQk eIXeAuUEi1LTUyvSMnOA4QuTluDgURLhTQdJ8xYXJOYWZ6ZDpE4xKkqJ8xqDJARAEhmleXBts Ji5xCgrJczLCHSIEE9BalFuZgmq/CtGcQ5GJWFeG5ApPJl5JXDTXwEtZgJavD0gG2RxSSJCSq qBce6ULtO1K+raMmP53px663lFUKbUwSji+dlbBh/CS5SeFa//U8G71LB07pE3T/odzZV6T4T c/uP9eJFH6I4ZRS1nLwTmGS5e+t5n1wOVBy+mHbmVs2bGo/yLZ5gXbU862x4b+FB10Z6YQ1v9 FV8lfmqu/Lnm0ankHWm6Rx8xHr6n4iO8OID7qJESS3FGoqEWc1FxIgC7BLzTlAIAAA== X-Env-Sender: prvs=1701003ad=Andrew.Cooper3@citrix.com X-Msg-Ref: server-6.tower-31.messagelabs.com!1483533980!52605894!4 X-Originating-IP: [66.165.176.89] X-SpamReason: No, hits=0.0 required=7.0 tests=sa_preprocessor: VHJ1c3RlZCBJUDogNjYuMTY1LjE3Ni44OSA9PiAyMDMwMDc=\n, received_headers: No Received headers X-StarScan-Received: X-StarScan-Version: 9.1.1; banners=-,-,- X-VirusChecked: Checked Received: (qmail 46183 invoked from network); 4 Jan 2017 12:46:24 -0000 Received: from smtp.citrix.com (HELO SMTP.CITRIX.COM) (66.165.176.89) by server-6.tower-31.messagelabs.com with RC4-SHA encrypted SMTP; 4 Jan 2017 12:46:24 -0000 X-IronPort-AV: E=Sophos;i="5.33,459,1477958400"; d="scan'208";a="397866977" From: Andrew Cooper To: Xen-devel Date: Wed, 4 Jan 2017 12:39:39 +0000 Message-ID: <1483533584-8015-23-git-send-email-andrew.cooper3@citrix.com> X-Mailer: git-send-email 2.1.4 In-Reply-To: <1483533584-8015-1-git-send-email-andrew.cooper3@citrix.com> References: <1483533584-8015-1-git-send-email-andrew.cooper3@citrix.com> MIME-Version: 1.0 Cc: Andrew Cooper , Jan Beulich Subject: [Xen-devel] [PATCH 22/27] x86/cpuid: Perform max_leaf calculations in guest_cpuid() X-BeenThere: xen-devel@lists.xen.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: xen-devel-bounces@lists.xen.org Sender: "Xen-devel" X-Virus-Scanned: ClamAV using ClamSMTP Clamp the toolstack-providied max_leaf values in recalculate_cpuid_policy(), causing the per-domain policy to have guest-accurate data. Have guest_cpuid() exit early if a requested leaf is out of range, rather than falling into the legacy path. Signed-off-by: Andrew Cooper --- CC: Jan Beulich --- xen/arch/x86/cpuid.c | 36 ++++++++++++++++++++++++++++++++++++ xen/arch/x86/hvm/hvm.c | 21 --------------------- xen/arch/x86/traps.c | 23 ----------------------- xen/include/asm-x86/cpuid.h | 1 + 4 files changed, 37 insertions(+), 44 deletions(-) diff --git a/xen/arch/x86/cpuid.c b/xen/arch/x86/cpuid.c index 01bb906..d140482 100644 --- a/xen/arch/x86/cpuid.c +++ b/xen/arch/x86/cpuid.c @@ -279,6 +279,10 @@ void recalculate_cpuid_policy(struct domain *d) uint32_t fs[FSCAPINTS], max_fs[FSCAPINTS]; unsigned int i; + p->basic.max_leaf = min(p->basic.max_leaf, max->basic.max_leaf); + p->feat.max_subleaf = min(p->feat.max_subleaf, max->feat.max_subleaf); + p->extd.max_leaf = min(p->extd.max_leaf, max->extd.max_leaf); + cpuid_policy_to_featureset(p, fs); cpuid_policy_to_featureset(max, max_fs); @@ -306,6 +310,9 @@ void recalculate_cpuid_policy(struct domain *d) if ( !d->disable_migrate && !d->arch.vtsc ) __clear_bit(X86_FEATURE_ITSC, fs); + if ( p->basic.max_leaf < 0xd ) + __clear_bit(X86_FEATURE_XSAVE, fs); + /* Fold host's FDP_EXCP_ONLY and NO_FPU_SEL into guest's view. */ fs[FEATURESET_7b0] &= ~special_features[FEATURESET_7b0]; fs[FEATURESET_7b0] |= (host_policy.feat._7b0 & @@ -333,21 +340,50 @@ void guest_cpuid(const struct vcpu *v, unsigned int leaf, unsigned int subleaf, struct cpuid_leaf *res) { const struct domain *d = v->domain; + const struct cpuid_policy *p = d->arch.cpuid; *res = EMPTY_LEAF; /* * First pass: * - Dispatch the virtualised leaves to their respective handlers. + * - Perform max_leaf/subleaf calculations, maybe returning early. */ switch ( leaf ) { + case 0x0 ... 0x6: + case 0x8 ... 0xc: +#if 0 /* For when CPUID_GUEST_NR_BASIC isn't 0xd */ + case 0xe ... CPUID_GUEST_NR_BASIC - 1: +#endif + if ( leaf > p->basic.max_leaf ) + return; + break; + + case 0x7: + if ( subleaf > p->feat.max_subleaf ) + return; + break; + + case 0xd: + if ( subleaf > ARRAY_SIZE(p->xstate.raw) ) + return; + break; + case 0x40000000 ... 0x400000ff: if ( is_viridian_domain(d) ) return cpuid_viridian_leaves(v, leaf, subleaf, res); /* Fallthrough. */ case 0x40000100 ... 0x4fffffff: return cpuid_hypervisor_leaves(v, leaf, subleaf, res); + + case 0x80000000 ... 0x80000000 + CPUID_GUEST_NR_EXTD - 1: + if ( leaf > p->extd.max_leaf ) + return; + break; + + default: + return; } /* {pv,hvm}_cpuid() have this expectation. */ diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c index 7cda53f..1dd92e3 100644 --- a/xen/arch/x86/hvm/hvm.c +++ b/xen/arch/x86/hvm/hvm.c @@ -3305,27 +3305,6 @@ void hvm_cpuid(unsigned int input, unsigned int *eax, unsigned int *ebx, if ( !edx ) edx = &dummy; - if ( input & 0x7fffffff ) - { - /* - * Requests outside the supported leaf ranges return zero on AMD - * and the highest basic leaf output on Intel. Uniformly follow - * the AMD model as the more sane one. - */ - unsigned int limit; - - domain_cpuid(d, (input >> 16) != 0x8000 ? 0 : 0x80000000, 0, - &limit, &dummy, &dummy, &dummy); - if ( input > limit ) - { - *eax = 0; - *ebx = 0; - *ecx = 0; - *edx = 0; - return; - } - } - domain_cpuid(d, input, count, eax, ebx, ecx, edx); switch ( input ) diff --git a/xen/arch/x86/traps.c b/xen/arch/x86/traps.c index 1c384cf..aed96c3 100644 --- a/xen/arch/x86/traps.c +++ b/xen/arch/x86/traps.c @@ -1032,29 +1032,6 @@ void pv_cpuid(struct cpu_user_regs *regs) subleaf = c = regs->_ecx; d = regs->_edx; - if ( leaf & 0x7fffffff ) - { - /* - * Requests outside the supported leaf ranges return zero on AMD - * and the highest basic leaf output on Intel. Uniformly follow - * the AMD model as the more sane one. - */ - unsigned int limit = (leaf >> 16) != 0x8000 ? 0 : 0x80000000, dummy; - - if ( !is_control_domain(currd) && !is_hardware_domain(currd) ) - domain_cpuid(currd, limit, 0, &limit, &dummy, &dummy, &dummy); - else - limit = cpuid_eax(limit); - if ( leaf > limit ) - { - regs->rax = 0; - regs->rbx = 0; - regs->rcx = 0; - regs->rdx = 0; - return; - } - } - if ( !is_control_domain(currd) && !is_hardware_domain(currd) ) domain_cpuid(currd, leaf, subleaf, &a, &b, &c, &d); else diff --git a/xen/include/asm-x86/cpuid.h b/xen/include/asm-x86/cpuid.h index c621a6f..7363263 100644 --- a/xen/include/asm-x86/cpuid.h +++ b/xen/include/asm-x86/cpuid.h @@ -87,6 +87,7 @@ struct cpuid_policy * Per-domain objects: * * - Guest accurate: + * - max_{,sub}leaf * - All FEATURESET_* words * * Everything else should be considered inaccurate, and not necesserily 0.