From: Maxim Levitsky <mlevitsk@redhat.com>
To: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org (open list:X86 ARCHITECTURE (32-BIT
AND 64-BIT)), Jim Mattson <jmattson@google.com>,
Joerg Roedel <joro@8bytes.org>, Borislav Petkov <bp@alien8.de>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Wanpeng Li <wanpengli@tencent.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>, Ingo Molnar <mingo@redhat.com>,
x86@kernel.org (maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)),
Sean Christopherson <seanjc@google.com>,
Maxim Levitsky <mlevitsk@redhat.com>
Subject: [PATCH v2 7/8] KVM: SVM: call avic_vcpu_load/avic_vcpu_put when enabling/disabling AVIC
Date: Tue, 13 Jul 2021 17:20:22 +0300 [thread overview]
Message-ID: <20210713142023.106183-8-mlevitsk@redhat.com> (raw)
In-Reply-To: <20210713142023.106183-1-mlevitsk@redhat.com>
Currently it is possible to have the following scenario:
1. AVIC is disabled by svm_refresh_apicv_exec_ctrl
2. svm_vcpu_blocking calls avic_vcpu_put which does nothing
3. svm_vcpu_unblocking enables the AVIC (due to KVM_REQ_APICV_UPDATE)
and then calls avic_vcpu_load
4. warning is triggered in avic_vcpu_load since
AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK was never cleared
While it is possible to just remove the warning, it seems to be more robust
to fully disable/enable AVIC in svm_refresh_apicv_exec_ctrl by calling the
avic_vcpu_load/avic_vcpu_put
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
---
arch/x86/kvm/svm/avic.c | 43 ++++++++++++++++++++++-------------------
arch/x86/kvm/svm/svm.c | 8 ++++++--
arch/x86/kvm/x86.c | 10 ++++++++--
3 files changed, 37 insertions(+), 24 deletions(-)
diff --git a/arch/x86/kvm/svm/avic.c b/arch/x86/kvm/svm/avic.c
index 64d1e1b6a305..0e4370fe0868 100644
--- a/arch/x86/kvm/svm/avic.c
+++ b/arch/x86/kvm/svm/avic.c
@@ -80,6 +80,28 @@ enum avic_ipi_failure_cause {
AVIC_IPI_FAILURE_INVALID_BACKING_PAGE,
};
+
+static void __avic_set_running(struct kvm_vcpu *vcpu, bool is_run)
+{
+ if (is_run)
+ avic_vcpu_load(vcpu, vcpu->cpu);
+ else
+ avic_vcpu_put(vcpu);
+}
+
+/*
+ * This function is called during VCPU halt/unhalt.
+ */
+static void avic_set_running(struct kvm_vcpu *vcpu, bool is_run)
+{
+ struct vcpu_svm *svm = to_svm(vcpu);
+
+ svm->avic_is_running = is_run;
+
+ if (kvm_vcpu_apicv_active(vcpu))
+ __avic_set_running(vcpu, is_run);
+}
+
/* Note:
* This function is called from IOMMU driver to notify
* SVM to schedule in a particular vCPU of a particular VM.
@@ -667,6 +689,7 @@ void svm_refresh_apicv_exec_ctrl(struct kvm_vcpu *vcpu)
}
vmcb_mark_dirty(vmcb, VMCB_AVIC);
+ __avic_set_running(vcpu, activated);
svm_set_pi_irte_mode(vcpu, activated);
}
@@ -960,9 +983,6 @@ void avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
int h_physical_id = kvm_cpu_get_apicid(cpu);
struct vcpu_svm *svm = to_svm(vcpu);
- if (!kvm_vcpu_apicv_active(vcpu))
- return;
-
/*
* Since the host physical APIC id is 8 bits,
* we can support host APIC ID upto 255.
@@ -990,9 +1010,6 @@ void avic_vcpu_put(struct kvm_vcpu *vcpu)
u64 entry;
struct vcpu_svm *svm = to_svm(vcpu);
- if (!kvm_vcpu_apicv_active(vcpu))
- return;
-
entry = READ_ONCE(*(svm->avic_physical_id_cache));
if (entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK)
avic_update_iommu_vcpu_affinity(vcpu, -1, 0);
@@ -1001,20 +1018,6 @@ void avic_vcpu_put(struct kvm_vcpu *vcpu)
WRITE_ONCE(*(svm->avic_physical_id_cache), entry);
}
-/*
- * This function is called during VCPU halt/unhalt.
- */
-static void avic_set_running(struct kvm_vcpu *vcpu, bool is_run)
-{
- struct vcpu_svm *svm = to_svm(vcpu);
-
- svm->avic_is_running = is_run;
- if (is_run)
- avic_vcpu_load(vcpu, vcpu->cpu);
- else
- avic_vcpu_put(vcpu);
-}
-
void svm_vcpu_blocking(struct kvm_vcpu *vcpu)
{
avic_set_running(vcpu, false);
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index d14be8aba2d7..2c291dfa0e04 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -1505,12 +1505,16 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
sd->current_vmcb = svm->vmcb;
indirect_branch_prediction_barrier();
}
- avic_vcpu_load(vcpu, cpu);
+
+ if (kvm_vcpu_apicv_active(vcpu))
+ avic_vcpu_load(vcpu, cpu);
}
static void svm_vcpu_put(struct kvm_vcpu *vcpu)
{
- avic_vcpu_put(vcpu);
+ if (kvm_vcpu_apicv_active(vcpu))
+ avic_vcpu_put(vcpu);
+
svm_prepare_host_switch(vcpu);
++vcpu->stat.host_state_reload;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index a91e35b92447..213d332664e2 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -9177,12 +9177,18 @@ void kvm_make_scan_ioapic_request(struct kvm *kvm)
void kvm_vcpu_update_apicv(struct kvm_vcpu *vcpu)
{
+ bool activate;
+
if (!lapic_in_kernel(vcpu))
return;
mutex_lock(&vcpu->kvm->apicv_update_lock);
- vcpu->arch.apicv_active = kvm_apicv_activated(vcpu->kvm);
+ activate = kvm_apicv_activated(vcpu->kvm);
+ if (vcpu->arch.apicv_active == activate)
+ goto out;
+
+ vcpu->arch.apicv_active = activate;
kvm_apic_update_apicv(vcpu);
static_call(kvm_x86_refresh_apicv_exec_ctrl)(vcpu);
@@ -9194,7 +9200,7 @@ void kvm_vcpu_update_apicv(struct kvm_vcpu *vcpu)
*/
if (!vcpu->arch.apicv_active)
kvm_make_request(KVM_REQ_EVENT, vcpu);
-
+out:
mutex_unlock(&vcpu->kvm->apicv_update_lock);
}
EXPORT_SYMBOL_GPL(kvm_vcpu_update_apicv);
--
2.26.3
next prev parent reply other threads:[~2021-07-13 14:21 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-13 14:20 [PATCH v2 0/8] My AVIC patch queue Maxim Levitsky
2021-07-13 14:20 ` [PATCH v2 1/8] KVM: SVM: svm_set_vintr don't warn if AVIC is active but is about to be deactivated Maxim Levitsky
2021-07-13 14:20 ` [PATCH v2 2/8] KVM: SVM: tweak warning about enabled AVIC on nested entry Maxim Levitsky
2021-07-13 14:20 ` [PATCH v2 3/8] KVM: SVM: use vmcb01 in svm_refresh_apicv_exec_ctrl Maxim Levitsky
2021-07-13 14:20 ` [PATCH v2 4/8] KVM: x86: APICv: drop immediate APICv disablement on current vCPU Maxim Levitsky
2021-07-13 14:20 ` [PATCH v2 5/8] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM Maxim Levitsky
2021-07-26 22:34 ` Paolo Bonzini
2021-07-27 13:22 ` Maxim Levitsky
2021-07-13 14:20 ` [PATCH v2 6/8] KVM: SVM: add warning for mistmatch between AVIC state and AVIC access page state Maxim Levitsky
2021-07-13 14:20 ` Maxim Levitsky [this message]
2021-07-13 14:20 ` [PATCH v2 8/8] KVM: x86: hyper-v: Deactivate APICv only when AutoEOI feature is in use Maxim Levitsky
2021-07-18 12:13 ` Maxim Levitsky
2021-07-19 7:47 ` Vitaly Kuznetsov
2021-07-19 9:00 ` Maxim Levitsky
2021-07-19 9:23 ` Vitaly Kuznetsov
2021-07-19 9:58 ` Maxim Levitsky
2021-07-19 18:49 ` Sean Christopherson
2021-07-20 9:40 ` Maxim Levitsky
2021-07-22 9:12 ` KVM's support for non default APIC base Maxim Levitsky
2021-08-02 9:20 ` Maxim Levitsky
2021-08-06 21:55 ` Sean Christopherson
2021-08-09 9:40 ` Maxim Levitsky
2021-08-09 15:57 ` Sean Christopherson
2021-08-09 16:47 ` Jim Mattson
2021-08-10 20:42 ` Maxim Levitsky
2021-07-22 17:35 ` [PATCH v2 8/8] KVM: x86: hyper-v: Deactivate APICv only when AutoEOI feature is in use Maxim Levitsky
2021-07-22 19:06 ` Sean Christopherson
2021-07-27 13:05 ` Maxim Levitsky
2021-07-27 17:48 ` Ben Gardon
2021-07-27 18:17 ` Sean Christopherson
2021-07-29 14:10 ` Maxim Levitsky
2021-07-26 17:24 ` [PATCH v2 0/8] My AVIC patch queue Paolo Bonzini
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=20210713142023.106183-8-mlevitsk@redhat.com \
--to=mlevitsk@redhat.com \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=jmattson@google.com \
--cc=joro@8bytes.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=seanjc@google.com \
--cc=tglx@linutronix.de \
--cc=vkuznets@redhat.com \
--cc=wanpengli@tencent.com \
--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).