From: Sean Christopherson <sean.j.christopherson@intel.com>
To: Xiaoyao Li <xiaoyao.li@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
hpa@zytor.com, Paolo Bonzini <pbonzini@redhat.com>,
Andy Lutomirski <luto@kernel.org>,
tony.luck@intel.com, peterz@infradead.org, fenghua.yu@intel.com,
x86@kernel.org, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 6/8] kvm: vmx: Extend VMX's #AC interceptor to handle split lock #AC happens in guest
Date: Tue, 3 Mar 2020 11:08:26 -0800 [thread overview]
Message-ID: <20200303190826.GS1439@linux.intel.com> (raw)
In-Reply-To: <20200206070412.17400-7-xiaoyao.li@intel.com>
On Thu, Feb 06, 2020 at 03:04:10PM +0800, Xiaoyao Li wrote:
> There are two types of #AC can be generated in Intel CPUs:
> 1. legacy alignment check #AC;
> 2. split lock #AC;
>
> Legacy alignment check #AC can be injected to guest if guest has enabled
> alignemnet check.
>
> when host enables split lock detectin, i.e., split_lock_detect != off,
> there will be an unexpected #AC in guest and intercepted by KVM because
> KVM doesn't virtualize this feature to guest and hardware value of
> MSR_TEST_CTRL.SLD bit stays unchanged when vcpu is running.
>
> To handle this unexpected #AC, treat guest just like host usermode that
> calling handle_user_split_lock():
> - If host is sld_warn, it warns and set TIF_SLD so that __switch_to_xtra()
> does the MSR_TEST_CTRL.SLD bit switching when control transfer to/from
> this vcpu.
> - If host is sld_fatal, forward #AC to userspace, the similar as sending
> SIGBUS.
>
> Suggested-by: Sean Christopherson <sean.j.christopherson@intel.com>
> Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>
> ---
> v3:
> - Use handle_user_split_lock() to handle unexpected #AC in guest.
> ---
> arch/x86/kvm/vmx/vmx.c | 31 ++++++++++++++++++++++++++++---
> 1 file changed, 28 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> index c475fa2aaae0..822211975e6c 100644
> --- a/arch/x86/kvm/vmx/vmx.c
> +++ b/arch/x86/kvm/vmx/vmx.c
> @@ -4557,6 +4557,12 @@ static int handle_machine_check(struct kvm_vcpu *vcpu)
> return 1;
> }
>
> +static inline bool guest_cpu_alignment_check_enabled(struct kvm_vcpu *vcpu)
> +{
> + return vmx_get_cpl(vcpu) == 3 && kvm_read_cr0_bits(vcpu, X86_CR0_AM) &&
> + (kvm_get_rflags(vcpu) & X86_EFLAGS_AC);
> +}
> +
> static int handle_exception_nmi(struct kvm_vcpu *vcpu)
> {
> struct vcpu_vmx *vmx = to_vmx(vcpu);
> @@ -4622,9 +4628,6 @@ static int handle_exception_nmi(struct kvm_vcpu *vcpu)
> return handle_rmode_exception(vcpu, ex_no, error_code);
>
> switch (ex_no) {
> - case AC_VECTOR:
> - kvm_queue_exception_e(vcpu, AC_VECTOR, error_code);
> - return 1;
> case DB_VECTOR:
> dr6 = vmcs_readl(EXIT_QUALIFICATION);
> if (!(vcpu->guest_debug &
> @@ -4653,6 +4656,28 @@ static int handle_exception_nmi(struct kvm_vcpu *vcpu)
> kvm_run->debug.arch.pc = vmcs_readl(GUEST_CS_BASE) + rip;
> kvm_run->debug.arch.exception = ex_no;
> break;
> + case AC_VECTOR:
> + /*
> + * Inject #AC back to guest only when guest enables legacy
> + * alignment check.
The comment should call out that checking split_lock_detect_enabled() is an
optimization.
/*
* Reflect #AC to the guest if it's expecting the #AC, i.e. has
* legacy alignment check enabled. Pre-check host split lock
* support to avoid the VMREADs needed to check legacy #AC,
* i.e. reflect the #AC if the only possible source is legacy
* alignment checks.
*/
> + * Otherwise, it must be an unexpected split lock #AC of guest
> + * since hardware SPLIT_LOCK_DETECT bit keeps unchanged set
> + * when vcpu is running. In this case, treat guest the same as
> + * user space application that calls handle_user_split_lock():
> + * - If sld_state = sld_warn, it sets TIF_SLD and disables SLD
> + * for this vcpu thread.
> + * - If sld_state = sld_fatal, we forward #AC to userspace,
> + * similar as sending SIGBUS.
I'd prefer to avoid talking about sld_state at all and instead keep those
details in handle_user_split_lock().
> + */
> + if (!split_lock_detect_enabled() ||
> + guest_cpu_alignment_check_enabled(vcpu)) {
> + kvm_queue_exception_e(vcpu, AC_VECTOR, error_code);
> + return 1;
> + }
Something like:
/*
* Forward the #AC to userspace if kernel policy does not allow
* temporarily disabling split lock detection.
*/
> + if (handle_user_split_lock(kvm_rip_read(vcpu)))
> + return 1;
> + /* fall through */
> default:
> kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
> kvm_run->ex.exception = ex_no;
> --
> 2.23.0
>
next prev parent reply other threads:[~2020-03-03 19:08 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-06 7:04 [PATCH v3 0/8] kvm/split_lock: Add feature split lock detection support in kvm Xiaoyao Li
2020-02-06 7:04 ` [PATCH v3 1/8] x86/split_lock: Export handle_user_split_lock() Xiaoyao Li
2020-03-03 18:42 ` Sean Christopherson
2020-02-06 7:04 ` [PATCH v3 2/8] x86/split_lock: Ensure X86_FEATURE_SPLIT_LOCK_DETECT means the existence of feature Xiaoyao Li
2020-03-03 18:55 ` Sean Christopherson
2020-03-03 19:41 ` Sean Christopherson
2020-03-04 1:49 ` Xiaoyao Li
2020-03-05 16:23 ` Sean Christopherson
2020-03-06 2:15 ` Xiaoyao Li
2020-03-04 2:20 ` Xiaoyao Li
2020-02-06 7:04 ` [PATCH v3 3/8] x86/split_lock: Cache the value of MSR_TEST_CTRL in percpu data Xiaoyao Li
2020-02-06 20:23 ` Arvind Sankar
2020-02-07 4:18 ` Xiaoyao Li
2020-03-03 19:18 ` Sean Christopherson
2020-03-05 6:48 ` Xiaoyao Li
2020-02-06 7:04 ` [PATCH v3 4/8] x86/split_lock: Add and export split_lock_detect_enabled() and split_lock_detect_fatal() Xiaoyao Li
2020-03-03 18:59 ` Sean Christopherson
2020-02-06 7:04 ` [PATCH v3 5/8] kvm: x86: Emulate split-lock access as a write Xiaoyao Li
2020-02-06 7:04 ` [PATCH v3 6/8] kvm: vmx: Extend VMX's #AC interceptor to handle split lock #AC happens in guest Xiaoyao Li
2020-03-03 19:08 ` Sean Christopherson [this message]
2020-02-06 7:04 ` [PATCH v3 7/8] kvm: x86: Emulate MSR IA32_CORE_CAPABILITIES Xiaoyao Li
2020-02-06 7:04 ` [PATCH v3 8/8] x86: vmx: virtualize split lock detection Xiaoyao Li
2020-02-07 18:27 ` Arvind Sankar
2020-02-08 4:51 ` Xiaoyao Li
2020-03-03 19:30 ` Sean Christopherson
2020-03-05 14:16 ` Xiaoyao Li
2020-03-05 16:49 ` Sean Christopherson
2020-03-06 0:29 ` Xiaoyao Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200303190826.GS1439@linux.intel.com \
--to=sean.j.christopherson@intel.com \
--cc=bp@alien8.de \
--cc=fenghua.yu@intel.com \
--cc=hpa@zytor.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
--cc=x86@kernel.org \
--cc=xiaoyao.li@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).