From: Paolo Bonzini <pbonzini@redhat.com>
To: "Kang, Luwei" <luwei.kang@intel.com>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Cc: "rkrcmar@redhat.com" <rkrcmar@redhat.com>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"mingo@redhat.com" <mingo@redhat.com>,
"hpa@zytor.com" <hpa@zytor.com>,
"x86@kernel.org" <x86@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Chao Peng <chao.p.peng@linux.intel.com>
Subject: Re: [patch v2 3/8] KVM: x86: add Intel processor trace virtualization mode
Date: Tue, 14 Nov 2017 11:47:22 +0100 [thread overview]
Message-ID: <de621c4c-6116-2340-428b-ad3e03151752@redhat.com> (raw)
In-Reply-To: <82D7661F83C1A047AF7DC287873BF1E167E37638@SHSMSX101.ccr.corp.intel.com>
On 14/11/2017 08:31, Kang, Luwei wrote:
>>> + if (!(_cpu_based_2nd_exec_control & SECONDARY_EXEC_PT_USE_GPA) ||
>>> + !(_vmexit_control & VM_EXIT_CLEAR_IA32_RTIT_CTL) ||
>>> + !(_vmentry_control & VM_ENTRY_LOAD_IA32_RTIT_CTL)) {
>>> + _cpu_based_2nd_exec_control &= ~SECONDARY_EXEC_PT_USE_GPA;
>>
>> Also, you are not checking anywhere if the SUPPRESS_PIP controls are available. This is probably the best place.
>
> SUPPRESS_PIP(should be "CONCEAL", will fix it.) is use for control of processor trace packet.
> I think we should clear it when in SYSTEM mode (For example, PIPs are
> generated on VM exit, with NonRoot=0. On VM exit to SMM, VMCS packets
> are additionally generated). Why need check this here?
I agree. But if SUPPRESS_PIP/CONCEAL is not available, you cannot
enable HOST_GUEST mode I think?
>>
>>> + _vmexit_control &= ~VM_EXIT_CLEAR_IA32_RTIT_CTL;
>>> + _vmentry_control &= ~VM_ENTRY_LOAD_IA32_RTIT_CTL;
>>
>> These two are not needed; disabling SECONDARY_EXEC_PT_USE_GPA is enough.
>> The tracing mode will revert to PT_SYSTEM, which does not use the load/clear RTIT_CTL controls.
>
> The status of *_RTIT_CTL should be same with SECONDARY_EXEC_PT_USE_GPA
> or would cause VM-entry failed.
> (architecture-instruction-set-extensions-programming-reference 5.2.3)
Indeed:
- disable "use GPA" here
- hardware_setup notices !cpu_has_vmx_pt_use_gpa() and switches tracing
mode to system
- vmx_vcpu_setup (via vmx_vmentry_control and vmx_vmexit_control) turns
off the IA32_RIT_CTL vmentry/vmexit controls (and also turns off
SUPPRESS_PIP)
- everybody is happy
Paolo
next prev parent reply other threads:[~2017-11-14 10:47 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-30 22:05 [patch v2 0/8] Intel Processor Trace virtulization enabling Luwei Kang
2017-10-30 22:05 ` [patch v2 1/8] perf/x86/intel/pt: Move Intel-PT MSR bit definitions to a public header Luwei Kang
2017-10-30 22:05 ` [patch v2 2/8] perf/x86/intel/pt: change pt_cap_get() to a public function Luwei Kang
2017-10-30 22:05 ` [patch v2 3/8] KVM: x86: add Intel processor trace virtualization mode Luwei Kang
2017-11-13 16:29 ` Paolo Bonzini
2017-11-14 7:11 ` Kang, Luwei
2017-11-13 16:32 ` Paolo Bonzini
2017-11-14 7:31 ` Kang, Luwei
2017-11-14 10:47 ` Paolo Bonzini [this message]
2017-10-30 22:05 ` [patch v2 4/8] KVM: x86: add Intel processor trace cpuid emulataion Luwei Kang
2017-11-13 15:36 ` Paolo Bonzini
2017-11-14 2:56 ` Kang, Luwei
2017-10-30 22:05 ` [patch v2 5/8] KVM: x86: add Intel processor trace context for each vcpu Luwei Kang
2017-10-30 22:05 ` [patch v2 6/8] KVM: x86: Implement Intel processor trace context switch Luwei Kang
2017-11-13 16:01 ` Paolo Bonzini
2017-10-30 22:05 ` [patch v2 7/8] KVM: x86: add Intel PT msr RTIT_CTL read/write Luwei Kang
2017-11-13 16:22 ` Paolo Bonzini
2017-11-14 6:59 ` Kang, Luwei
2017-11-14 10:34 ` Paolo Bonzini
2017-10-30 22:05 ` [patch v2 8/8] KVM: x86: Disable intercept for Intel processor trace MSRs Luwei Kang
2017-11-13 16:24 ` Paolo Bonzini
2017-11-14 7:01 ` Kang, Luwei
2017-11-02 14:57 ` [patch v2 0/8] Intel Processor Trace virtulization enabling Paolo Bonzini
2017-11-03 6:09 ` Kang, Luwei
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=de621c4c-6116-2340-428b-ad3e03151752@redhat.com \
--to=pbonzini@redhat.com \
--cc=chao.p.peng@linux.intel.com \
--cc=hpa@zytor.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luwei.kang@intel.com \
--cc=mingo@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/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).