From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754315AbcFPLT0 (ORCPT ); Thu, 16 Jun 2016 07:19:26 -0400 Received: from mx1.redhat.com ([209.132.183.28]:37814 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754293AbcFPLTX (ORCPT ); Thu, 16 Jun 2016 07:19:23 -0400 Subject: Re: [PATCH v2 2/3] KVM: VMX: validate individual bits of guest MSR_IA32_FEATURE_CONTROL To: kvm@vger.kernel.org, rkrcmar@redhat.com, Thomas Gleixner , Ingo Molnar , "H . Peter Anvin" , x86@kernel.org, linux-kernel@vger.kernel.org, Gleb Natapov , Boris Petkov , Tony Luck , Andi Kleen , Ashok Raj References: <20160616060531.30028-1-haozhong.zhang@intel.com> <20160616060531.30028-3-haozhong.zhang@intel.com> <088f9fcb-a648-05d4-530a-790cb084bd09@redhat.com> <20160616111655.ltcgle2cju2fksrx@hz-desktop> From: Paolo Bonzini Message-ID: Date: Thu, 16 Jun 2016 13:19:16 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.1.0 MIME-Version: 1.0 In-Reply-To: <20160616111655.ltcgle2cju2fksrx@hz-desktop> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.39]); Thu, 16 Jun 2016 11:19:22 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 16/06/2016 13:16, Haozhong Zhang wrote: >> However, I think FEATURE_CONTROL_LOCKED should always be writable. If >> you change that, it's simpler to just do |= and &= in the caller. > > These two functions (add/del) are to prevent callers from forgetting > setting/clearing FEATURE_CONTROL_LOCKED in > msr_ia32_feature_control_valid_bits: it should be set if any feature > bit is set, and be cleared if all feature bits are cleared. The second > rule could relaxed as we can always present MSR_IA32_FEATURE_CONTROL > to guest. Yes, this means that FEATURE_CONTROL_LOCKED effectively is always valid. So you end up with just &= to clear and |= to set. > I'm okey to let callers take care for the locked bit. > >>> + to_vmx(vcpu)->msr_ia32_feature_control_valid_bits |= >>> + bits | FEATURE_CONTROL_LOCKED; >>> +} >>> + >>> +static void feature_control_valid_bits_del(struct kvm_vcpu *vcpu, uint64_t bits) >>> +{ >>> + uint64_t *valid_bits = >>> + &to_vmx(vcpu)->msr_ia32_feature_control_valid_bits; >>> + ASSERT(!(bits & ~FEATURE_CONTROL_MAX_VALID_BITS)); >>> + *valid_bits &= ~bits; >>> + if (!(*valid_bits & ~FEATURE_CONTROL_LOCKED)) >>> + *valid_bits = 0; >>> +} >>> + >>> #define VMCS12_OFFSET(x) offsetof(struct vmcs12, x) >>> #define FIELD(number, name) [number] = VMCS12_OFFSET(name) >>> #define FIELD64(number, name) [number] = VMCS12_OFFSET(name), \ >>> @@ -2864,6 +2897,14 @@ static int vmx_get_vmx_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) >>> return 0; >>> } >>> >>> +static inline bool vmx_feature_control_msr_valid(struct kvm_vcpu *vcpu, >>> + uint64_t val) >>> +{ >>> + uint64_t valid_bits = to_vmx(vcpu)->msr_ia32_feature_control_valid_bits; >>> + >>> + return valid_bits && !(val & ~valid_bits); >>> +} >>> /* >>> * Reads an msr value (of 'msr_index') into 'pdata'. >>> * Returns 0 on success, non-0 otherwise. >>> @@ -2906,7 +2947,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) >>> msr_info->data = vmcs_read64(GUEST_BNDCFGS); >>> break; >>> case MSR_IA32_FEATURE_CONTROL: >>> - if (!nested_vmx_allowed(vcpu)) >>> + if (!vmx_feature_control_msr_valid(vcpu, 0)) >> >> You can remove this if completely in patch 1. It's okay to make the MSR >> always available. >> > > But then it also allows all bits to be set by guests, even though some > features are not available. Note that this is "get". Of course the "if" must stay in vmx_set_msr. Paolo