| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Fri, 08 Mar 2019 14:10:30 +0800
From: Xiaoyao Li <xiaoyao.li@...ux.intel.com>
To: Paolo Bonzini <pbonzini@...hat.com>,
Fenghua Yu <fenghua.yu@...el.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
H Peter Anvin <hpa@...or.com>,
Dave Hansen <dave.hansen@...el.com>,
Ashok Raj <ashok.raj@...el.com>,
Peter Zijlstra <peterz@...radead.org>,
Ravi V Shankar <ravi.v.shankar@...el.com>
Cc: linux-kernel <linux-kernel@...r.kernel.org>, x86 <x86@...nel.org>,
kvm@...r.kernel.org
Subject: Re: [PATCH v4 16/17] kvm: x86: Add support IA32_CORE_CAPABILITY MSR
Hi, Paolo
On Mon, 2019-03-04 at 09:42 +0100, Paolo Bonzini wrote:
> On 02/03/19 03:45, Fenghua Yu wrote:
> > From: Xiaoyao Li <xiaoyao.li@...ux.intel.com>
> >
> > MSR IA32_CORE_CAPABILITY is a feature-enumerating MSR, bit 5 of which
> > reports the capability of enabling detection of split locks (will be
> > supported on future processors based on Tremont microarchitecture and
> > later).
> >
> > Please check the latest Intel Architecture Instruction Set Extensions
> > and Future Features Programming Reference for more detailed information
> > on the MSR and the split lock bit.
> >
> > 1. Expose it to user space as a feature-enumerating MSR, so that user
> > space can query it.
> >
> > 2. Emualte MSR_IA32_CORE_CAPABILITY with vmx->core_capability. And add the
> > get and set handler of MSR_IA32_CORE_CAPABILITY.
> > For uesrspace, it can set this MSR when customizing features of guest,
> > also it can read the value of this MSR of guest.
> > For guest, as it's a feature-enumerating MSR, guest only can read it.
> >
> > Signed-off-by: Xiaoyao Li <xiaoyao.li@...ux.intel.com>
> > Signed-off-by: Fenghua Yu <fenghua.yu@...el.com>
> > ---
> > arch/x86/include/asm/kvm_host.h | 1 +
> > arch/x86/kvm/vmx/vmx.c | 23 +++++++++++++++++++++++
> > arch/x86/kvm/vmx/vmx.h | 1 +
> > arch/x86/kvm/x86.c | 17 ++++++++++++++++-
> > 4 files changed, 41 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/x86/include/asm/kvm_host.h
> > b/arch/x86/include/asm/kvm_host.h
> > index 180373360e34..208f15570d17 100644
> > --- a/arch/x86/include/asm/kvm_host.h
> > +++ b/arch/x86/include/asm/kvm_host.h
> > @@ -1527,6 +1527,7 @@ int kvm_pv_send_ipi(struct kvm *kvm, unsigned long
> > ipi_bitmap_low,
> > unsigned long icr, int op_64_bit);
> >
> > u64 kvm_get_arch_capabilities(void);
> > +u64 kvm_get_core_capability(void);
> > void kvm_define_shared_msr(unsigned index, u32 msr);
> > int kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
> >
> > diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> > index 30a6bcd735ec..3e03c6e1e558 100644
> > --- a/arch/x86/kvm/vmx/vmx.c
> > +++ b/arch/x86/kvm/vmx/vmx.c
> > @@ -1679,6 +1679,12 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct
> > msr_data *msr_info)
> >
> > msr_info->data = to_vmx(vcpu)->spec_ctrl;
> > break;
> > + case MSR_IA32_CORE_CAPABILITY:
> > + if (!msr_info->host_initiated &&
> > + !guest_cpuid_has(vcpu, X86_FEATURE_CORE_CAPABILITY))
> > + return 1;
> > + msr_info->data = vmx->core_capability;
> > + break;
> > case MSR_IA32_ARCH_CAPABILITIES:
> > if (!msr_info->host_initiated &&
> > !guest_cpuid_has(vcpu, X86_FEATURE_ARCH_CAPABILITIES))
> > @@ -1891,6 +1897,21 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct
> > msr_data *msr_info)
> > vmx_disable_intercept_for_msr(vmx->vmcs01.msr_bitmap,
> > MSR_IA32_PRED_CMD,
> > MSR_TYPE_W);
> > break;
> > + case MSR_IA32_CORE_CAPABILITY:
> > + if (!msr_info->host_initiated)
> > + return 1;
> > + if (data & ~CORE_CAP_SPLIT_LOCK_DETECT)
> > + return 1;
> > +
> > + /*
> > + * Since AC split lock is a hardware feature, and there is no
> > + * software emulation yet, we cannot enable it for guest if
> > + * host hardware doesn't support it.
> > + */
> > + if (!boot_cpu_has(X86_FEATURE_SPLIT_LOCK_DETECT))
> > + data &= ~CORE_CAP_SPLIT_LOCK_DETECT;
> > + vmx->core_capability = data;
> > + break;
> > case MSR_IA32_ARCH_CAPABILITIES:
> > if (!msr_info->host_initiated)
> > return 1;
> > @@ -4083,6 +4104,8 @@ static void vmx_vcpu_setup(struct vcpu_vmx *vmx)
> > ++vmx->nmsrs;
> > }
> >
> > + vmx->core_capability = kvm_get_core_capability();
> > +
> > vmx->arch_capabilities = kvm_get_arch_capabilities();
> >
> > vm_exit_controls_init(vmx, vmx_vmexit_ctrl());
> > diff --git a/arch/x86/kvm/vmx/vmx.h b/arch/x86/kvm/vmx/vmx.h
> > index 0ac0a64c7790..cc22379991f3 100644
> > --- a/arch/x86/kvm/vmx/vmx.h
> > +++ b/arch/x86/kvm/vmx/vmx.h
> > @@ -191,6 +191,7 @@ struct vcpu_vmx {
> > u64 msr_guest_kernel_gs_base;
> > #endif
> >
> > + u64 core_capability;
> > u64 arch_capabilities;
> > u64 spec_ctrl;
> >
> > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> > index 941f932373d0..c3c9e3f2d08a 100644
> > --- a/arch/x86/kvm/x86.c
> > +++ b/arch/x86/kvm/x86.c
> > @@ -1125,7 +1125,8 @@ static u32 msrs_to_save[] = {
> > #endif
> > MSR_IA32_TSC, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA,
> > MSR_IA32_FEATURE_CONTROL, MSR_IA32_BNDCFGS, MSR_TSC_AUX,
> > - MSR_IA32_SPEC_CTRL, MSR_IA32_ARCH_CAPABILITIES,
> > + MSR_IA32_SPEC_CTRL, MSR_IA32_CORE_CAPABILITY,
> > + MSR_IA32_ARCH_CAPABILITIES,
> > MSR_IA32_RTIT_CTL, MSR_IA32_RTIT_STATUS, MSR_IA32_RTIT_CR3_MATCH,
> > MSR_IA32_RTIT_OUTPUT_BASE, MSR_IA32_RTIT_OUTPUT_MASK,
> > MSR_IA32_RTIT_ADDR0_A, MSR_IA32_RTIT_ADDR0_B,
> > @@ -1197,11 +1198,22 @@ static u32 msr_based_features[] = {
> >
> > MSR_F10H_DECFG,
> > MSR_IA32_UCODE_REV,
> > + MSR_IA32_CORE_CAPABILITY,
> > MSR_IA32_ARCH_CAPABILITIES,
> > };
> >
> > static unsigned int num_msr_based_features;
> >
> > +u64 kvm_get_core_capability(void)
> > +{
> > + u64 data;
> > +
> > + rdmsrl_safe(MSR_IA32_CORE_CAPABILITY, &data);
>
> This patch should be merged with the previous patch. Also here you
> should add:
>
> data &= CORE_CAP_SPLIT_LOCK_DETECT;
I agree with this.
> so that non-virtualizable features are hidden and
>
> if (boot_cpu_has(X86_FEATURE_SPLIT_LOCK_DETECT))
> data |= CORE_CAP_SPLIT_LOCK_DETECT;
>
> so that userspace gets "for free" the FMS list that will be added
> later to the kernel.
I think it's redundant. Because there is no case that
rdmsrl_safe(MSR_IA32_CORE_CAPABILITY, &data) shows no split lock detection while
boot_cpu_has(X86_FEATURE_SPLIT_LOCK_DETECT).
Xiaoyao
> Thanks,
>
> Paolo
>
> > +
> > + return data;
> > +}
> > +EXPORT_SYMBOL_GPL(kvm_get_core_capability);
> > +
> > u64 kvm_get_arch_capabilities(void)
> > {
> > u64 data;
> > @@ -1227,6 +1239,9 @@ EXPORT_SYMBOL_GPL(kvm_get_arch_capabilities);
> > static int kvm_get_msr_feature(struct kvm_msr_entry *msr)
> > {
> > switch (msr->index) {
> > + case MSR_IA32_CORE_CAPABILITY:
> > + msr->data = kvm_get_core_capability();
> > + break;
> > case MSR_IA32_ARCH_CAPABILITIES:
> > msr->data = kvm_get_arch_capabilities();
> > break;
> >
>
>
Powered by blists - more mailing lists