qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Subject: [PULL 20/30] i386: Implement pseudo 'hv-avic' ('hv-apicv') enlightenment
Date: Sun,  3 Oct 2021 09:42:40 +0200	[thread overview]
Message-ID: <20211003074250.60869-21-pbonzini@redhat.com> (raw)
In-Reply-To: <20211003074250.60869-1-pbonzini@redhat.com>

From: Vitaly Kuznetsov <vkuznets@redhat.com>

The enlightenment allows to use Hyper-V SynIC with hardware APICv/AVIC
enabled. Normally, Hyper-V SynIC disables these hardware features and
suggests the guest to use paravirtualized AutoEOI feature. Linux-4.15
gains support for conditional APICv/AVIC disablement, the feature
stays on until the guest tries to use AutoEOI feature with SynIC. With
'HV_DEPRECATING_AEOI_RECOMMENDED' bit exposed, modern enough Windows/
Hyper-V versions should follow the recommendation and not use the
(unwanted) feature.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210902093530.345756-7-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 docs/hyperv.txt                | 10 +++++++++-
 target/i386/cpu.c              |  4 ++++
 target/i386/cpu.h              |  1 +
 target/i386/kvm/hyperv-proto.h |  1 +
 target/i386/kvm/kvm.c          | 10 +++++++++-
 5 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/docs/hyperv.txt b/docs/hyperv.txt
index 072709a68f..cd1ea3bbe9 100644
--- a/docs/hyperv.txt
+++ b/docs/hyperv.txt
@@ -189,7 +189,15 @@ enabled.
 
 Requires: hv-vpindex, hv-synic, hv-time, hv-stimer
 
-3.17. hv-no-nonarch-coresharing=on/off/auto
+3.18. hv-avic (hv-apicv)
+=======================
+The enlightenment allows to use Hyper-V SynIC with hardware APICv/AVIC enabled.
+Normally, Hyper-V SynIC disables these hardware feature and suggests the guest
+to use paravirtualized AutoEOI feature.
+Note: enabling this feature on old hardware (without APICv/AVIC support) may
+have negative effect on guest's performace.
+
+3.19. hv-no-nonarch-coresharing=on/off/auto
 ===========================================
 This enlightenment tells guest OS that virtual processors will never share a
 physical core unless they are reported as sibling SMT threads. This information
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 2a19eba56d..8154343cc4 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -6644,6 +6644,8 @@ static void x86_cpu_initfn(Object *obj)
     object_property_add_alias(obj, "sse4_1", obj, "sse4.1");
     object_property_add_alias(obj, "sse4_2", obj, "sse4.2");
 
+    object_property_add_alias(obj, "hv-apicv", obj, "hv-avic");
+
     if (xcc->model) {
         x86_cpu_load_model(cpu, xcc->model);
     }
@@ -6831,6 +6833,8 @@ static Property x86_cpu_properties[] = {
                       HYPERV_FEAT_IPI, 0),
     DEFINE_PROP_BIT64("hv-stimer-direct", X86CPU, hyperv_features,
                       HYPERV_FEAT_STIMER_DIRECT, 0),
+    DEFINE_PROP_BIT64("hv-avic", X86CPU, hyperv_features,
+                      HYPERV_FEAT_AVIC, 0),
     DEFINE_PROP_ON_OFF_AUTO("hv-no-nonarch-coresharing", X86CPU,
                             hyperv_no_nonarch_cs, ON_OFF_AUTO_OFF),
     DEFINE_PROP_BOOL("hv-passthrough", X86CPU, hyperv_passthrough, false),
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 8a7209bbf2..65f0ee2caf 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1056,6 +1056,7 @@ typedef uint64_t FeatureWordArray[FEATURE_WORDS];
 #define HYPERV_FEAT_EVMCS               12
 #define HYPERV_FEAT_IPI                 13
 #define HYPERV_FEAT_STIMER_DIRECT       14
+#define HYPERV_FEAT_AVIC                15
 
 #ifndef HYPERV_SPINLOCK_NEVER_NOTIFY
 #define HYPERV_SPINLOCK_NEVER_NOTIFY             0xFFFFFFFF
diff --git a/target/i386/kvm/hyperv-proto.h b/target/i386/kvm/hyperv-proto.h
index 5fbb385cc1..89f81afda7 100644
--- a/target/i386/kvm/hyperv-proto.h
+++ b/target/i386/kvm/hyperv-proto.h
@@ -66,6 +66,7 @@
 #define HV_APIC_ACCESS_RECOMMENDED          (1u << 3)
 #define HV_SYSTEM_RESET_RECOMMENDED         (1u << 4)
 #define HV_RELAXED_TIMING_RECOMMENDED       (1u << 5)
+#define HV_DEPRECATING_AEOI_RECOMMENDED     (1u << 9)
 #define HV_CLUSTER_IPI_RECOMMENDED          (1u << 10)
 #define HV_EX_PROCESSOR_MASKS_RECOMMENDED   (1u << 11)
 #define HV_ENLIGHTENED_VMCS_RECOMMENDED     (1u << 14)
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index a9a8f77df3..68faf72e34 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -924,6 +924,13 @@ static struct {
         },
         .dependencies = BIT(HYPERV_FEAT_STIMER)
     },
+    [HYPERV_FEAT_AVIC] = {
+        .desc = "AVIC/APICv support (hv-avic/hv-apicv)",
+        .flags = {
+            {.func = HV_CPUID_ENLIGHTMENT_INFO, .reg = R_EAX,
+             .bits = HV_DEPRECATING_AEOI_RECOMMENDED}
+        }
+    },
 };
 
 static struct kvm_cpuid2 *try_get_hv_cpuid(CPUState *cs, int max,
@@ -1373,7 +1380,8 @@ static int hyperv_fill_cpuids(CPUState *cs,
     c->eax = hv_build_cpuid_leaf(cs, HV_CPUID_ENLIGHTMENT_INFO, R_EAX);
     c->ebx = cpu->hyperv_spinlock_attempts;
 
-    if (hyperv_feat_enabled(cpu, HYPERV_FEAT_VAPIC)) {
+    if (hyperv_feat_enabled(cpu, HYPERV_FEAT_VAPIC) &&
+        !hyperv_feat_enabled(cpu, HYPERV_FEAT_AVIC)) {
         c->eax |= HV_APIC_ACCESS_RECOMMENDED;
     }
 
-- 
2.31.1




  parent reply	other threads:[~2021-10-03  8:12 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-03  7:42 [PULL 00/30] Misc changes for 2021-10-03 Paolo Bonzini
2021-10-03  7:42 ` [PULL 01/30] qapi/machine: Fix an incorrect comment of SMPConfiguration Paolo Bonzini
2021-10-03  7:42 ` [PULL 02/30] machine: Deprecate "parameter=0" SMP configurations Paolo Bonzini
2021-10-03  7:42 ` [PULL 03/30] machine: Minor refactor/fix for the smp parsers Paolo Bonzini
2021-10-03  7:42 ` [PULL 04/30] machine: Uniformly use maxcpus to calculate the omitted parameters Paolo Bonzini
2021-10-03  7:42 ` [PULL 05/30] machine: Set the value of cpus to match maxcpus if it's omitted Paolo Bonzini
2021-10-03  7:42 ` [PULL 06/30] machine: Improve the error reporting of smp parsing Paolo Bonzini
2021-10-03  7:42 ` [PULL 07/30] qtest/numa-test: Use detailed -smp CLIs in pc_dynamic_cpu_cfg Paolo Bonzini
2021-10-03  7:42 ` [PULL 08/30] qtest/numa-test: Use detailed -smp CLIs in test_def_cpu_split Paolo Bonzini
2021-10-03  7:42 ` [PULL 09/30] machine: Prefer cores over sockets in smp parsing since 6.2 Paolo Bonzini
2021-10-03  7:42 ` [PULL 10/30] machine: Use ms instead of global current_machine in sanity-check Paolo Bonzini
2021-10-03  7:42 ` [PULL 11/30] machine: Tweak the order of topology members in struct CpuTopology Paolo Bonzini
2021-10-03  7:42 ` [PULL 12/30] machine: Make smp_parse generic enough for all arches Paolo Bonzini
2021-10-03  7:42 ` [PULL 13/30] machine: Remove smp_parse callback from MachineClass Paolo Bonzini
2021-10-03  7:42 ` [PULL 14/30] machine: Move smp_prefer_sockets to struct SMPCompatProps Paolo Bonzini
2021-10-03  7:42 ` [PULL 15/30] machine: Use g_autoptr in machine_set_smp Paolo Bonzini
2021-10-03  7:42 ` [PULL 16/30] machine: Put all sanity-check in the generic SMP parser Paolo Bonzini
2021-10-03  7:42 ` [PULL 17/30] i386: Support KVM_CAP_ENFORCE_PV_FEATURE_CPUID Paolo Bonzini
2021-10-03  7:42 ` [PULL 18/30] i386: Support KVM_CAP_HYPERV_ENFORCE_CPUID Paolo Bonzini
2021-10-03  7:42 ` [PULL 19/30] i386: Move HV_APIC_ACCESS_RECOMMENDED bit setting to hyperv_fill_cpuids() Paolo Bonzini
2021-10-03  7:42 ` Paolo Bonzini [this message]
2021-10-03  7:42 ` [PULL 21/30] i386: Make Hyper-V version id configurable Paolo Bonzini
2021-10-03  7:42 ` [PULL 22/30] i386: Change the default Hyper-V version to match WS2016 Paolo Bonzini
2021-10-03  7:42 ` [PULL 23/30] configure: Loosen GCC requirement from 7.5.0 to 7.4.0 Paolo Bonzini
2021-10-03  7:42 ` [PULL 24/30] virtio-mem-pci: Fix memory leak when creating MEMORY_DEVICE_SIZE_CHANGE event Paolo Bonzini
2021-10-03  7:42 ` [PULL 25/30] qapi: Include qom-path in MEMORY_DEVICE_SIZE_CHANGE qapi events Paolo Bonzini
2021-10-03  7:42 ` [PULL 26/30] monitor: Rate-limit MEMORY_DEVICE_SIZE_CHANGE qapi events per device Paolo Bonzini
2021-10-03  7:42 ` [PULL 27/30] tpm: mark correct memory region range dirty when clearing RAM Paolo Bonzini
2021-10-03  7:42 ` [PULL 28/30] softmmu/memory_mapping: never merge ranges accross memory regions Paolo Bonzini
2021-10-03  7:42 ` [PULL 29/30] softmmu/memory_mapping: factor out adding physical memory ranges Paolo Bonzini
2021-10-03  7:42 ` [PULL 30/30] softmmu/memory_mapping: optimize for RamDiscardManager sections Paolo Bonzini
2021-10-03 14:44 ` [PULL 00/30] Misc changes for 2021-10-03 Philippe Mathieu-Daudé
2021-10-03 15:12 ` Richard Henderson

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=20211003074250.60869-21-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=vkuznets@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 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).