From: Vitaly Kuznetsov <vkuznets@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>, kvm@vger.kernel.org
Cc: Sean Christopherson <seanjc@google.com>,
Wanpeng Li <wanpengli@tencent.com>,
Jim Mattson <jmattson@google.com>,
Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [PATCH v2 6/4] selftests: kvm: Add basic Hyper-V clocksources tests
Date: Thu, 18 Mar 2021 16:23:35 +0100 [thread overview]
Message-ID: <87zgz05vuw.fsf@vitty.brq.redhat.com> (raw)
In-Reply-To: <41897f61-9d1a-519b-e1cb-e19efa34ab55@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 1315 bytes --]
Paolo Bonzini <pbonzini@redhat.com> writes:
> On 18/03/21 15:52, Vitaly Kuznetsov wrote:
>> Paolo Bonzini <pbonzini@redhat.com> writes:
>>
>>> On 18/03/21 15:09, Vitaly Kuznetsov wrote:
>>>> +static inline void check_tsc_msr_tsc_page(struct ms_hyperv_tsc_page *tsc_page)
>>>> +{
>>>> + u64 r1, r2, t1, t2;
>>>> + s64 delta_ns;
>>>> +
>>>> + /* Compare TSC page clocksource with HV_X64_MSR_TIME_REF_COUNT */
>>>> + t1 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
>>>> + r1 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
>>>> + nop_loop();
>>>> + t2 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
>>>> + r2 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
>>>> +
>>>> + delta_ns = ((r2 - r1) - (t2 - t1)) * 100;
>>>> + if (delta_ns < 0)
>>>> + delta_ns = -delta_ns;
>>>> +
>>>> + /* 1% tolerance */
>>>> + GUEST_ASSERT(delta_ns * 100 < (t2 - t1) * 100);
>>>> +}
>>>> +
>>>
>>> I think you should also be able to check r1 and r2 individually, not
>>> just r1 and r2. Is that correct?
>>
>> Right, we could've checked r1 == t1 and r2 == t2 actually (with some
>> tiny margin of course). Let me try that.
>
> np, I can do that too. Just checking my recollection of the TLFS.
>
I already hacked it, patch attached :-) Feel free to squash it or put
your own version in.
--
Vitaly
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-selftests-KVM-hyper-v-check_tsc_msr_tsc_page-fix.patch --]
[-- Type: text/x-patch, Size: 1846 bytes --]
From dae56d5d5c5fc7442d4dabf96a9c322acb42f458 Mon Sep 17 00:00:00 2001
From: Vitaly Kuznetsov <vkuznets@redhat.com>
Date: Thu, 18 Mar 2021 16:21:18 +0100
Subject: [PATCH] selftests: KVM: hyper-v: check_tsc_msr_tsc_page() fix.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
---
.../selftests/kvm/x86_64/hyperv_clock.c | 25 ++++++++++---------
1 file changed, 13 insertions(+), 12 deletions(-)
diff --git a/tools/testing/selftests/kvm/x86_64/hyperv_clock.c b/tools/testing/selftests/kvm/x86_64/hyperv_clock.c
index 39d6491d8458..aacb6d2697da 100644
--- a/tools/testing/selftests/kvm/x86_64/hyperv_clock.c
+++ b/tools/testing/selftests/kvm/x86_64/hyperv_clock.c
@@ -82,22 +82,23 @@ static inline void check_tsc_msr_rdtsc(void)
static inline void check_tsc_msr_tsc_page(struct ms_hyperv_tsc_page *tsc_page)
{
- u64 r1, r2, t1, t2;
- s64 delta_ns;
+ u64 t1, t2;
- /* Compare TSC page clocksource with HV_X64_MSR_TIME_REF_COUNT */
- t1 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
- r1 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
+ /*
+ * Make TSC run for a while to make sure clocksources don't diverge
+ * over time.
+ */
nop_loop();
- t2 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
- r2 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
- delta_ns = ((r2 - r1) - (t2 - t1)) * 100;
- if (delta_ns < 0)
- delta_ns = -delta_ns;
+ /*
+ * Get readings from TSC page and Reference TSC clocksources and make
+ * sure they match.
+ */
+ t1 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
+ t2 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
- /* 1% tolerance */
- GUEST_ASSERT(delta_ns * 100 < (t2 - t1) * 100);
+ /* 10us tolerance */
+ GUEST_ASSERT(t2 - t1 <= 100);
}
static void guest_main(struct ms_hyperv_tsc_page *tsc_page, vm_paddr_t tsc_page_gpa)
--
2.30.2
next prev parent reply other threads:[~2021-03-18 15:24 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-16 14:37 [PATCH v2 0/4] KVM: x86: hyper-v: TSC page fixes Vitaly Kuznetsov
2021-03-16 14:37 ` [PATCH v2 1/4] KVM: x86: hyper-v: Limit guest to writing zero to HV_X64_MSR_TSC_EMULATION_STATUS Vitaly Kuznetsov
2021-03-16 14:37 ` [PATCH v2 2/4] KVM: x86: hyper-v: Prevent using not-yet-updated TSC page by secondary CPUs Vitaly Kuznetsov
2021-03-18 17:02 ` Marcelo Tosatti
2021-03-18 18:04 ` Marcelo Tosatti
2021-03-18 18:05 ` Paolo Bonzini
2021-03-18 18:30 ` Marcelo Tosatti
2021-03-19 9:29 ` Vitaly Kuznetsov
2021-03-16 14:37 ` [PATCH v2 3/4] KVM: x86: hyper-v: Track Hyper-V TSC page status Vitaly Kuznetsov
2021-03-17 8:07 ` Paolo Bonzini
2021-03-17 11:19 ` [PATCH v2 5/4] KVM: x86: hyper-v: Briefly document enum hv_tsc_page_status Vitaly Kuznetsov
2021-03-16 14:37 ` [PATCH v2 4/4] KVM: x86: hyper-v: Don't touch TSC page values when guest opted for re-enlightenment Vitaly Kuznetsov
2021-03-18 14:09 ` [PATCH v2 6/4] selftests: kvm: Add basic Hyper-V clocksources tests Vitaly Kuznetsov
2021-03-18 14:26 ` Paolo Bonzini
2021-03-18 14:52 ` Vitaly Kuznetsov
2021-03-18 15:01 ` Paolo Bonzini
2021-03-18 15:23 ` Vitaly Kuznetsov [this message]
2021-03-18 15:27 ` Paolo Bonzini
2021-03-18 16:57 ` Marcelo Tosatti
2021-03-18 17:50 ` Paolo Bonzini
2021-03-18 17:55 ` Marcelo Tosatti
2021-03-19 9:35 ` Vitaly Kuznetsov
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=87zgz05vuw.fsf@vitty.brq.redhat.com \
--to=vkuznets@redhat.com \
--cc=jmattson@google.com \
--cc=kvm@vger.kernel.org \
--cc=mtosatti@redhat.com \
--cc=pbonzini@redhat.com \
--cc=seanjc@google.com \
--cc=wanpengli@tencent.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).