All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Radim Krčmář" <rkrcmar@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: kvm@vger.kernel.org, Paolo Bonzini <pbonzini@redhat.com>,
	Nadav Amit <namit@cs.technion.ac.il>,
	Gleb Natapov <gleb@kernel.org>
Subject: [PATCH 6/8] KVM: x86: allow mixed APIC mode broadcast
Date: Thu, 29 Jan 2015 22:48:53 +0100	[thread overview]
Message-ID: <1422568135-28402-7-git-send-email-rkrcmar@redhat.com> (raw)
In-Reply-To: <1422568135-28402-1-git-send-email-rkrcmar@redhat.com>

Broadcast allowed only one global APIC mode, but mixed modes are
theoretically possible.  x2APIC IPI doesn't mean 0xff as broadcast,
the rest does.

(We also save one rcu dereference with broadcasts.)

Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  2 +-
 arch/x86/kvm/lapic.c            | 13 ++++++++-----
 2 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 843bea0e70fd..26d0f0f646d3 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -556,7 +556,7 @@ struct kvm_apic_map {
 	struct rcu_head rcu;
 	u8 ldr_bits;
 	/* fields bellow are used to decode ldr values in different modes */
-	u32 cid_shift, cid_mask, lid_mask, broadcast;
+	u32 cid_shift, cid_mask, lid_mask;
 	struct kvm_lapic *phys_map[256];
 	/* first index is cluster id second is cpu id in a cluster */
 	struct kvm_lapic *logical_map[16][16];
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 871ce6a2485b..fab007509047 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -151,7 +151,6 @@ static void recalculate_apic_map(struct kvm *kvm)
 	new->cid_shift = 8;
 	new->cid_mask = 0;
 	new->lid_mask = 0xff;
-	new->broadcast = APIC_BROADCAST;
 
 	kvm_for_each_vcpu(i, vcpu, kvm) {
 		struct kvm_lapic *apic = vcpu->arch.apic;
@@ -163,7 +162,6 @@ static void recalculate_apic_map(struct kvm *kvm)
 			new->ldr_bits = 32;
 			new->cid_shift = 16;
 			new->cid_mask = new->lid_mask = 0xffff;
-			new->broadcast = X2APIC_BROADCAST;
 		} else if (kvm_apic_get_reg(apic, APIC_LDR)) {
 			if (kvm_apic_get_reg(apic, APIC_DFR) ==
 							APIC_DFR_CLUSTER) {
@@ -683,6 +681,7 @@ bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src,
 	struct kvm_lapic **dst;
 	int i;
 	bool ret = false;
+	bool x2apic_ipi = src && apic_x2apic_mode(src);
 
 	*r = -1;
 
@@ -694,15 +693,19 @@ bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src,
 	if (irq->shorthand)
 		return false;
 
+	/* XXX: A superset of x2APIC broadcasts is fine in practice as long as
+	 * we don't support APIC ID > 0xfeffffff.
+	 */
+	if ((x2apic_ipi ? GET_APIC_DEST_FIELD(irq->dest_id) : irq->dest_id)
+	    == APIC_BROADCAST)
+		return false;
+
 	rcu_read_lock();
 	map = rcu_dereference(kvm->arch.apic_map);
 
 	if (!map)
 		goto out;
 
-	if (irq->dest_id == map->broadcast)
-		goto out;
-
 	ret = true;
 
 	if (irq->dest_mode == APIC_DEST_PHYSICAL) {
-- 
2.2.2


  parent reply	other threads:[~2015-01-29 21:49 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-29 21:48 [PATCH 0/8] KVM: minor APIC fixes and cleanups Radim Krčmář
2015-01-29 21:48 ` [PATCH 1/8] KVM: x86: return bool from kvm_apic_match*() Radim Krčmář
2015-01-29 22:10   ` Joe Perches
2015-01-30  8:50     ` Paolo Bonzini
2015-01-29 21:48 ` [PATCH 2/8] KVM: x86: cleanup kvm_apic_match_*() Radim Krčmář
2015-01-30  8:52   ` Paolo Bonzini
2015-01-30 13:06     ` Radim Krčmář
2015-02-02 14:26     ` Radim Krčmář
2015-02-02 14:28       ` Paolo Bonzini
2015-02-02 14:30         ` Radim Krčmář
2015-02-02 14:29       ` Radim Krčmář
2015-01-29 21:48 ` [PATCH 3/8] KVM: x86: replace 0 with APIC_DEST_PHYSICAL Radim Krčmář
2015-01-29 21:48 ` [PATCH 4/8] KVM: x86: fix x2apic logical address matching Radim Krčmář
2015-01-29 21:48 ` [PATCH 5/8] KVM: x86: use MDA for interrupt matching Radim Krčmář
2015-01-30  9:03   ` Paolo Bonzini
2015-01-30 13:09     ` Radim Krčmář
2015-01-29 21:48 ` Radim Krčmář [this message]
2015-01-29 21:48 ` [PATCH 7/8] KVM: x86: avoid logical_map when it is invalid Radim Krčmář
2015-01-30  9:19   ` Paolo Bonzini
2015-01-30 14:21     ` Radim Krčmář
2015-01-30 14:21       ` Paolo Bonzini
2015-01-30  9:38   ` Paolo Bonzini
2015-01-30 14:56     ` Radim Krčmář
2015-01-30 15:10       ` Paolo Bonzini
2015-01-30 17:09         ` Radim Krčmář
2015-01-29 21:48 ` [PATCH 8/8] KVM: x86: simplify kvm_apic_map Radim Krčmář
2015-01-30  9:18   ` Paolo Bonzini
2015-01-30 15:14     ` Radim Krčmář
2015-01-30 15:23       ` Paolo Bonzini
2015-01-30 16:57         ` Radim Krčmář
2015-01-30 21:15           ` Paolo Bonzini
2015-01-30  9:22 ` [PATCH 0/8] KVM: minor APIC fixes and cleanups Paolo Bonzini
2015-01-30 15:20   ` Radim Krčmář
2015-01-30 15:24     ` 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=1422568135-28402-7-git-send-email-rkrcmar@redhat.com \
    --to=rkrcmar@redhat.com \
    --cc=gleb@kernel.org \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=namit@cs.technion.ac.il \
    --cc=pbonzini@redhat.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.