From: Sean Christopherson <seanjc@google.com>
To: Shivam Kumar <shivam.kumar1@nutanix.com>
Cc: pbonzini@redhat.com, maz@kernel.org, james.morse@arm.com,
borntraeger@linux.ibm.com, david@redhat.com, kvm@vger.kernel.org,
Shaju Abraham <shaju.abraham@nutanix.com>,
Manish Mishra <manish.mishra@nutanix.com>,
Anurag Madnawat <anurag.madnawat@nutanix.com>
Subject: Re: [PATCH v6 2/5] KVM: x86: Dirty quota-based throttling of vcpus
Date: Wed, 19 Oct 2022 15:42:10 +0000 [thread overview]
Message-ID: <Y1Aa0jQWOop2OHtC@google.com> (raw)
In-Reply-To: <a835d5c3-8742-e8f7-e810-a69a139c4349@nutanix.com>
On Tue, Oct 18, 2022, Shivam Kumar wrote:
> > > #ifdef CONFIG_HAVE_KVM_DIRTY_QUOTA
> > > static inline bool kvm_check_dirty_quota_request(struct kvm_vcpu *vcpu)
> > > {
> > > struct kvm_run *run = vcpu->run;
> > >
> > > run->exit_reason = KVM_EXIT_DIRTY_QUOTA_EXHAUSTED;
> > > run->dirty_quota_exit.count = vcpu->stat.generic.pages_dirtied;
> > > run->dirty_quota_exit.quota = READ_ONCE(run->dirty_quota);
> > >
> > > /*
> > > * Re-check the quota and exit if and only if the vCPU still
> > > exceeds its
> > > * quota. If userspace increases (or disables entirely) the
> > > quota, then
> > > * no exit is required as KVM is still honoring its ABI, e.g.
> > > userspace
> > > * won't even be aware that KVM temporarily detected an
> > > exhausted quota.
> > > */
> > > return run->dirty_quota_exit.count >= run->dirty_quota_exit.quota;
> > > }
> > > #endif
> > >
> > > And then arch usage is simply something like:
> > >
> > > if (kvm_check_dirty_quota_request(vcpu)) {
> > > r = 0;
> > > goto out;
> > > }
> > If we are not even checking for request KVM_REQ_DIRTY_QUOTA_EXIT, what's
> > the use of kvm_make_request in patch 1?
> Ok, so we don't need to explicitely check for request here because we are
> checking the quota directly but we do need to make the request when the
> quota is exhausted so as to guarantee that we enter the if block "if
> (kvm_request_pending(vcpu))".
>
> Please let me know if my understanding is correct or if I am missing
> something.
The request needs to be explicitly checked, I simply unintentionally omitted that
code in the above snippet. kvm_check_request() also _clears_ the request, which
is very important as not clearing the request will prevent KVM from entering the
guest.
The intent is to check KVM_REQ_DIRTY_QUOTA_EXIT in kvm_check_dirty_quota_request().
>
> Also, should I add ifdef here:
>
> #ifdef CONFIG_HAVE_KVM_DIRTY_QUOTA
> if (kvm_check_dirty_quota_request(vcpu)) {
> r = 0;
> goto out;
> }
> #endif
No need for an #ifdef in the caller, the call is from arch code and architectures
that don't "select HAVE_KVM_DIRTY_QUOTA" shouldn't be checking for a dirty quota
exit.
next prev parent reply other threads:[~2022-10-19 15:48 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-15 10:10 [PATCH v6 0/5] KVM: Dirty quota-based throttling Shivam Kumar
2022-09-15 10:10 ` [PATCH v6 1/5] KVM: Implement dirty quota-based throttling of vcpus Shivam Kumar
2022-09-15 13:21 ` Christian Borntraeger
2022-09-15 14:34 ` Christian Borntraeger
2022-10-07 18:18 ` Sean Christopherson
2022-10-09 18:36 ` Shivam Kumar
2022-10-10 6:12 ` Christian Borntraeger
2022-10-07 19:08 ` Sean Christopherson
2022-10-07 19:20 ` Sean Christopherson
2022-10-09 18:49 ` Shivam Kumar
2022-10-10 16:09 ` Sean Christopherson
2022-10-09 19:30 ` Shivam Kumar
2022-10-10 5:41 ` Shivam Kumar
2022-10-17 5:28 ` Shivam Kumar
2022-10-19 16:01 ` Sean Christopherson
2022-09-15 10:10 ` [PATCH v6 2/5] KVM: x86: Dirty " Shivam Kumar
2022-10-07 19:30 ` Sean Christopherson
2022-10-09 19:05 ` Shivam Kumar
2022-10-18 17:43 ` Shivam Kumar
2022-10-19 15:42 ` Sean Christopherson [this message]
2022-10-09 19:17 ` Shivam Kumar
2022-09-15 10:10 ` [PATCH v6 3/5] KVM: arm64: " Shivam Kumar
2022-09-15 10:10 ` [PATCH v6 4/5] KVM: s390x: " Shivam Kumar
2022-09-15 13:24 ` Christian Borntraeger
2022-09-15 10:10 ` [PATCH v6 5/5] KVM: selftests: Add selftests for dirty quota throttling Shivam Kumar
2022-10-07 18:29 ` Sean Christopherson
2022-10-09 19:26 ` Shivam Kumar
2022-10-10 15:47 ` Sean Christopherson
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=Y1Aa0jQWOop2OHtC@google.com \
--to=seanjc@google.com \
--cc=anurag.madnawat@nutanix.com \
--cc=borntraeger@linux.ibm.com \
--cc=david@redhat.com \
--cc=james.morse@arm.com \
--cc=kvm@vger.kernel.org \
--cc=manish.mishra@nutanix.com \
--cc=maz@kernel.org \
--cc=pbonzini@redhat.com \
--cc=shaju.abraham@nutanix.com \
--cc=shivam.kumar1@nutanix.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).