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: [Qemu-devel] [PULL 01/47] i386/kvm: add support for Hyper-V TLB flush
Date: Sat, 30 Jun 2018 08:09:54 +0200	[thread overview]
Message-ID: <20180630061040.6018-2-pbonzini@redhat.com> (raw)
In-Reply-To: <20180630061040.6018-1-pbonzini@redhat.com>

From: Vitaly Kuznetsov <vkuznets@redhat.com>

Add support for Hyper-V TLB flush which recently got added to KVM.

Just like regular Hyper-V we announce HV_EX_PROCESSOR_MASKS_RECOMMENDED
regardless of how many vCPUs we have. Windows is 'smart' and uses less
expensive non-EX Hypercall whenever possible (when it wants to flush TLB
for all vCPUs or the maximum vCPU index in the vCPU set requires flushing
is less than 64).

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20180610184927.19309-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 target/i386/cpu.c          |  1 +
 target/i386/cpu.h          |  1 +
 target/i386/hyperv-proto.h |  1 +
 target/i386/kvm.c          | 15 ++++++++++++++-
 4 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 1e6a7d0..d57f310 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -5415,6 +5415,7 @@ static Property x86_cpu_properties[] = {
     DEFINE_PROP_BOOL("hv-stimer", X86CPU, hyperv_stimer, false),
     DEFINE_PROP_BOOL("hv-frequencies", X86CPU, hyperv_frequencies, false),
     DEFINE_PROP_BOOL("hv-reenlightenment", X86CPU, hyperv_reenlightenment, false),
+    DEFINE_PROP_BOOL("hv-tlbflush", X86CPU, hyperv_tlbflush, false),
     DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true),
     DEFINE_PROP_BOOL("enforce", X86CPU, enforce_cpuid, false),
     DEFINE_PROP_BOOL("kvm", X86CPU, expose_kvm, true),
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 8eaefee..8b1ddfe 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1367,6 +1367,7 @@ struct X86CPU {
     bool hyperv_stimer;
     bool hyperv_frequencies;
     bool hyperv_reenlightenment;
+    bool hyperv_tlbflush;
     bool check_cpuid;
     bool enforce_cpuid;
     bool expose_kvm;
diff --git a/target/i386/hyperv-proto.h b/target/i386/hyperv-proto.h
index 93352eb..d6d5a79 100644
--- a/target/i386/hyperv-proto.h
+++ b/target/i386/hyperv-proto.h
@@ -58,6 +58,7 @@
 #define HV_APIC_ACCESS_RECOMMENDED          (1u << 3)
 #define HV_SYSTEM_RESET_RECOMMENDED         (1u << 4)
 #define HV_RELAXED_TIMING_RECOMMENDED       (1u << 5)
+#define HV_EX_PROCESSOR_MASKS_RECOMMENDED   (1u << 11)
 
 /*
  * Basic virtualized MSRs
diff --git a/target/i386/kvm.c b/target/i386/kvm.c
index 032f0ad..ebb2d23 100644
--- a/target/i386/kvm.c
+++ b/target/i386/kvm.c
@@ -601,7 +601,8 @@ static bool hyperv_enabled(X86CPU *cpu)
             cpu->hyperv_runtime ||
             cpu->hyperv_synic ||
             cpu->hyperv_stimer ||
-            cpu->hyperv_reenlightenment);
+            cpu->hyperv_reenlightenment ||
+            cpu->hyperv_tlbflush);
 }
 
 static int kvm_arch_set_tsc_khz(CPUState *cs)
@@ -839,6 +840,18 @@ int kvm_arch_init_vcpu(CPUState *cs)
         if (cpu->hyperv_vapic) {
             c->eax |= HV_APIC_ACCESS_RECOMMENDED;
         }
+        if (cpu->hyperv_tlbflush) {
+            if (kvm_check_extension(cs->kvm_state,
+                                    KVM_CAP_HYPERV_TLBFLUSH) <= 0) {
+                fprintf(stderr, "Hyper-V TLB flush support "
+                        "(requested by 'hv-tlbflush' cpu flag) "
+                        " is not supported by kernel\n");
+                return -ENOSYS;
+            }
+            c->eax |= HV_REMOTE_TLB_FLUSH_RECOMMENDED;
+            c->eax |= HV_EX_PROCESSOR_MASKS_RECOMMENDED;
+        }
+
         c->ebx = cpu->hyperv_spinlock_attempts;
 
         c = &cpuid_data.entries[cpuid_i++];
-- 
1.8.3.1

  reply	other threads:[~2018-06-30  6:10 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-30  6:09 [Qemu-devel] [PULL 00/47] More misc patches for 3.0 soft freeze Paolo Bonzini
2018-06-30  6:09 ` Paolo Bonzini [this message]
2018-06-30  6:09 ` [Qemu-devel] [PULL 02/47] configure: add sanity check to catch builds from "git archive" Paolo Bonzini
2018-06-30  6:09 ` [Qemu-devel] [PULL 03/47] include: Add IEC binary prefixes in "qemu/units.h" Paolo Bonzini
2018-06-30  6:09 ` [Qemu-devel] [PULL 04/47] vdi: Use definitions from "qemu/units.h" Paolo Bonzini
2018-06-30  6:09 ` [Qemu-devel] [PULL 05/47] x86/cpu: " Paolo Bonzini
2018-06-30  6:09 ` [Qemu-devel] [PULL 06/47] checkpatch: Recognize IEC binary prefix definitions Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 07/47] hw: Use IEC binary prefix definitions from "qemu/units.h" Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 08/47] hw: Directly use "qemu/units.h" instead of "qemu/cutils.h" Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 09/47] hw/ivshmem: Use the IEC binary prefix definitions Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 10/47] hw/ipack: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 11/47] hw/scsi: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 12/47] hw/smbios: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 13/47] hw/xen: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 14/47] hw/tpm: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 15/47] hw/block: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 16/47] hw/display: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 17/47] hw/misc: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 18/47] hw/riscv: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 19/47] hw/m68k: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 20/47] hw/sparc: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 21/47] hw/s390x: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 22/47] hw/hppa: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 23/47] hw/xtensa: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 24/47] hw/alpha: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 25/47] hw/tricore: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 26/47] hw/microblaze: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 27/47] hw/nios2: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 28/47] hw/cris: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 29/47] hw/lm32: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 30/47] hw/sh4: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 31/47] hw/mips/r4k: Constify params_size Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 32/47] hw/mips: Use the IEC binary prefix definitions Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 33/47] hw/ppc: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 34/47] hw/i386: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 35/47] hw/net: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 36/47] hw/usb: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 37/47] hw/sd: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 38/47] hw/vfio: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 39/47] hw/virtio: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 40/47] hw/rdma: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 41/47] cutils: Do not include "qemu/units.h" directly Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 42/47] monitor: Use the IEC binary prefix definitions Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 43/47] vl: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 44/47] tests/crypto: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 45/47] linux-user: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 46/47] bsd-user: " Paolo Bonzini
2018-06-30  6:10 ` [Qemu-devel] [PULL 47/47] serial: Open non-block Paolo Bonzini
2018-06-30 15:39 ` [Qemu-devel] [PULL 00/47] More misc patches for 3.0 soft freeze Peter Maydell
2018-06-30 15:51   ` Paolo Bonzini
2018-06-30 16:50     ` [Qemu-devel] [PATCH] !fixup 052f529eb3d07170b18b8d0920bc8c450e389a2f Philippe Mathieu-Daudé
2018-07-02  8:37       ` Paolo Bonzini
2018-07-02 13:32       ` 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=20180630061040.6018-2-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).