All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Alejandro Jimenez <alejandro.j.jimenez@oracle.com>,
	Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>,
	Maxim Levitsky <mlevitsk@redhat.com>,
	Li RongQing <lirongqing@baidu.com>
Subject: [PATCH v4 07/32] KVM: x86: Don't inhibit APICv/AVIC if xAPIC ID mismatch is due to 32-bit ID
Date: Sat,  1 Oct 2022 00:58:50 +0000	[thread overview]
Message-ID: <20221001005915.2041642-8-seanjc@google.com> (raw)
In-Reply-To: <20221001005915.2041642-1-seanjc@google.com>

Truncate the vcpu_id, a.k.a. x2APIC ID, to an 8-bit value when comparing
it against the xAPIC ID to avoid false positives (sort of) on systems
with >255 CPUs, i.e. with IDs that don't fit into a u8.  The intent of
APIC_ID_MODIFIED is to inhibit APICv/AVIC when the xAPIC is changed from
it's original value,

The mismatch isn't technically a false positive, as architecturally the
xAPIC IDs do end up being aliased in this scenario, and neither APICv
nor AVIC correctly handles IPI virtualization when there is aliasing.
However, KVM already deliberately does not honor the aliasing behavior
that results when an x2APIC ID gets truncated to an xAPIC ID.  I.e. the
resulting APICv/AVIC behavior is aligned with KVM's existing behavior
when KVM's x2APIC hotplug hack is effectively enabled.

If/when KVM provides a way to disable the hotplug hack, APICv/AVIC can
piggyback whatever logic disables the optimized APIC map (which is what
provides the hotplug hack), i.e. so that KVM's optimized map and APIC
virtualization yield the same behavior.

For now, fix the immediate problem of APIC virtualization being disabled
for large VMs, which is a much more pressing issue than ensuring KVM
honors architectural behavior for APIC ID aliasing.

Fixes: 3743c2f02517 ("KVM: x86: inhibit APICv/AVIC on changes to APIC ID or APIC base")
Reported-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/lapic.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 251856ba0750..2503f162eb95 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -2078,7 +2078,12 @@ static void kvm_lapic_xapic_id_updated(struct kvm_lapic *apic)
 	if (KVM_BUG_ON(apic_x2apic_mode(apic), kvm))
 		return;
 
-	if (kvm_xapic_id(apic) == apic->vcpu->vcpu_id)
+	/*
+	 * Deliberately truncate the vCPU ID when detecting a modified APIC ID
+	 * to avoid false positives if the vCPU ID, i.e. x2APIC ID, is a 32-bit
+	 * value.
+	 */
+	if (kvm_xapic_id(apic) == (u8)apic->vcpu->vcpu_id)
 		return;
 
 	kvm_set_apicv_inhibit(apic->vcpu->kvm, APICV_INHIBIT_REASON_APIC_ID_MODIFIED);
-- 
2.38.0.rc1.362.ged0d419d3c-goog


  parent reply	other threads:[~2022-10-01  1:00 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-01  0:58 [PATCH v4 00/32] KVM: x86: AVIC and local APIC fixes+cleanups Sean Christopherson
2022-10-01  0:58 ` [PATCH v4 01/32] KVM: x86: Blindly get current x2APIC reg value on "nodecode write" traps Sean Christopherson
2022-12-08 21:47   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 02/32] KVM: x86: Purge "highest ISR" cache when updating APICv state Sean Christopherson
2022-12-08 21:47   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 03/32] KVM: SVM: Flush the "current" TLB when activating AVIC Sean Christopherson
     [not found]   ` <b9f336f17eec6bfbb8429700e0f135d19813c576.camel@redhat.com>
2022-12-08 21:52     ` Maxim Levitsky
2022-12-09  0:40       ` Sean Christopherson
2022-12-08 22:02     ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 04/32] KVM: SVM: Process ICR on AVIC IPI delivery failure due to invalid target Sean Christopherson
2022-10-01  0:58 ` [PATCH v4 05/32] KVM: x86: Don't inhibit APICv/AVIC on xAPIC ID "change" if APIC is disabled Sean Christopherson
2022-12-08 21:53   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 06/32] KVM: x86: Track xAPIC ID only on userspace SET, _after_ vAPIC is updated Sean Christopherson
2022-12-08 21:53   ` Maxim Levitsky
2022-10-01  0:58 ` Sean Christopherson [this message]
2022-12-08 21:53   ` [PATCH v4 07/32] KVM: x86: Don't inhibit APICv/AVIC if xAPIC ID mismatch is due to 32-bit ID Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 08/32] KVM: SVM: Don't put/load AVIC when setting virtual APIC mode Sean Christopherson
2022-12-08 21:53   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 09/32] KVM: x86: Handle APICv updates for APIC "mode" changes via request Sean Christopherson
2022-12-08 21:54   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 10/32] KVM: x86: Move APIC access page helper to common x86 code Sean Christopherson
2022-12-08 21:55   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 11/32] KVM: x86: Inhibit APIC memslot if x2APIC and AVIC are enabled Sean Christopherson
2022-12-08 21:56   ` Maxim Levitsky
2022-12-16 19:03     ` Sean Christopherson
2022-12-16 19:40       ` Sean Christopherson
2022-12-27 11:25         ` Paolo Bonzini
2023-01-03 16:30           ` Sean Christopherson
2022-10-01  0:58 ` [PATCH v4 12/32] KVM: SVM: Replace "avic_mode" enum with "x2avic_enabled" boolean Sean Christopherson
2022-10-01  0:58 ` [PATCH v4 13/32] KVM: SVM: Compute dest based on sender's x2APIC status for AVIC kick Sean Christopherson
2022-10-01  0:58 ` [PATCH v4 14/32] KVM: SVM: Fix x2APIC Logical ID calculation for avic_kick_target_vcpus_fast Sean Christopherson
2022-10-01  0:58 ` [PATCH v4 15/32] Revert "KVM: SVM: Use target APIC ID to complete x2AVIC IRQs when possible" Sean Christopherson
2022-12-08 21:56   ` Maxim Levitsky
2022-10-01  0:58 ` [PATCH v4 16/32] KVM: SVM: Document that vCPU ID == APIC ID in AVIC kick fastpatch Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 17/32] KVM: SVM: Add helper to perform final AVIC "kick" of single vCPU Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 18/32] KVM: x86: Explicitly skip optimized logical map setup if vCPU's LDR==0 Sean Christopherson
2022-12-08 21:56   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 19/32] KVM: x86: Explicitly track all possibilities for APIC map's logical modes Sean Christopherson
2022-12-08 21:57   ` Maxim Levitsky
2022-12-16 18:39     ` Sean Christopherson
2022-12-16 23:34       ` Sean Christopherson
2022-12-27 11:30         ` Paolo Bonzini
2022-10-01  0:59 ` [PATCH v4 20/32] KVM: x86: Skip redundant x2APIC logical mode optimized cluster setup Sean Christopherson
2022-12-08 21:57   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 21/32] KVM: x86: Disable APIC logical map if logical ID covers multiple MDAs Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 22/32] KVM: x86: Disable APIC logical map if vCPUs are aliased in logical mode Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 23/32] KVM: x86: Honor architectural behavior for aliased 8-bit APIC IDs Sean Christopherson
2022-12-08 21:58   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 24/32] KVM: x86: Inhibit APICv/AVIC if the optimized physical map is disabled Sean Christopherson
2022-12-08 21:58   ` Maxim Levitsky
2022-12-09  0:56     ` Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 25/32] KVM: SVM: Inhibit AVIC if vCPUs are aliased in logical mode Sean Christopherson
2022-12-08 21:58   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 26/32] KVM: SVM: Always update local APIC on writes to logical dest register Sean Christopherson
2022-12-08 21:58   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 27/32] KVM: SVM: Update svm->ldr_reg cache even if LDR is "bad" Sean Christopherson
2022-12-08 21:59   ` Maxim Levitsky
2022-12-09  0:49     ` Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 28/32] KVM: SVM: Require logical ID to be power-of-2 for AVIC entry Sean Christopherson
2022-12-08 22:00   ` Maxim Levitsky
2022-12-29  8:27     ` mlevitsk
2023-01-04 10:08       ` Maxim Levitsky
2023-01-04 18:02       ` Sean Christopherson
2023-01-04 18:34         ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 29/32] KVM: SVM: Handle multiple logical targets in AVIC kick fastpath Sean Christopherson
2022-12-08 22:00   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 30/32] KVM: SVM: Ignore writes to Remote Read Data on AVIC write traps Sean Christopherson
2022-10-01  0:59 ` [PATCH v4 31/32] Revert "KVM: SVM: Do not throw warning when calling avic_vcpu_load on a running vcpu" Sean Christopherson
2022-12-08 22:01   ` Maxim Levitsky
2022-10-01  0:59 ` [PATCH v4 32/32] KVM: x86: Track required APICv inhibits with variable, not callback Sean Christopherson
2022-12-08 22:03   ` Maxim Levitsky
2022-12-27 11:22 ` [PATCH v4 00/32] KVM: x86: AVIC and local APIC fixes+cleanups 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=20221001005915.2041642-8-seanjc@google.com \
    --to=seanjc@google.com \
    --cc=alejandro.j.jimenez@oracle.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lirongqing@baidu.com \
    --cc=mlevitsk@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=suravee.suthikulpanit@amd.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.