From: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: pbonzini@redhat.com, rkrcmar@redhat.com, joro@8bytes.org,
        Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Subject: [PATCH v2 0/3] KVM: SVM: Fix guest not booting w/ AVIC enabled
Date: Tue, 12 Sep 2017 10:42:39 -0500
Message-Id: <1505230962-86304-1-git-send-email-suravee.suthikulpanit@amd.com>
MIME-Version: 1.0
Content-Type: text/plain
SpamDiagnosticOutput: 1:99
SpamDiagnosticMetadata: NSPM
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 12 Sep 2017 15:42:57.2325 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CY4PR12MB1733
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 1093
Lines: 26

Certain QEMU options fails to boot VM guest w/ SVM AVIC enabled
(e.g. modprobe kvm_amd avic=1). Investigation shows that this mainly
due to AVIC hardware does not trap into hypervisor when guest OS
writes to APIC_EOI register. 

The boot hang is caused by missing timer interrupt when using in-kernel
PIT model (e.g. launch qemu w/ '-no-hpet' option) since it requires
irq acknowledgmen before injecting another interrupt in case
irq re-injection is enabled (normally default).

Changes from V1 (https://lkml.org/lkml/2017/9/5/826)
 * Consolidate irqchip_split() check to only one place (per Radim). 

Suravee Suthikulpanit (3):
  KVM: SVM: Refactor AVIC vcpu initialization into avic_init_vcpu()
  KVM: Add struct kvm_vcpu pointer parameter to get_enable_apicv()
  KVM: SVM: Add irqchip_split() checks before enabling AVIC

 arch/x86/include/asm/kvm_host.h |  2 +-
 arch/x86/kvm/svm.c              | 43 ++++++++++++++++++++++++++++-------------
 arch/x86/kvm/vmx.c              |  2 +-
 arch/x86/kvm/x86.c              |  2 +-
 4 files changed, 33 insertions(+), 16 deletions(-)

-- 
1.8.3.1