From: Zhimin Feng <fengzhimin@bytedance.com>
To: x86@kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: pbonzini@redhat.com, seanjc@google.com, vkuznets@redhat.com,
wanpengli@tencent.com, jmattson@google.com, joro@8bytes.org,
tglx@linutronix.de, mingo@redhat.com, bp@alien8.de,
hpa@zytor.com, fweisbec@gmail.com, zhouyibo@bytedance.com,
zhanghaozhong@bytedance.com,
Zhimin Feng <fengzhimin@bytedance.com>
Subject: [RFC: timer passthrough 1/9] KVM: vmx: hook set_next_event for getting the host tscd
Date: Fri, 5 Feb 2021 18:03:09 +0800 [thread overview]
Message-ID: <20210205100317.24174-2-fengzhimin@bytedance.com> (raw)
In-Reply-To: <20210205100317.24174-1-fengzhimin@bytedance.com>
In order to get the host tscd value,
we need to hook set_next_event function
Signed-off-by: Zhimin Feng <fengzhimin@bytedance.com>
---
arch/x86/include/asm/kvm_host.h | 21 +++++++++++++++++
arch/x86/kvm/vmx/vmx.c | 51 +++++++++++++++++++++++++++++++++++++++++
arch/x86/kvm/x86.c | 1 +
kernel/time/tick-common.c | 1 +
4 files changed, 74 insertions(+)
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 7e5f33a0d0e2..eb6a611963b7 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -34,6 +34,7 @@
#include <asm/kvm_page_track.h>
#include <asm/kvm_vcpu_regs.h>
#include <asm/hyperv-tlfs.h>
+#include <linux/clockchips.h>
#define __KVM_HAVE_ARCH_VCPU_DEBUGFS
@@ -520,6 +521,24 @@ struct kvm_vcpu_hv {
cpumask_t tlb_flush;
};
+enum tick_device_mode {
+ TICKDEV_MODE_PERIODIC,
+ TICKDEV_MODE_ONESHOT,
+};
+
+struct tick_device {
+ struct clock_event_device *evtdev;
+ enum tick_device_mode mode;
+};
+
+struct timer_passth_info {
+ u64 host_tscd;
+ struct clock_event_device *curr_dev;
+
+ void (*orig_event_handler)(struct clock_event_device *dev);
+ int (*orig_set_next_event)(unsigned long evt, struct clock_event_device *dev);
+};
+
struct kvm_vcpu_arch {
/*
* rip and regs accesses must go through
@@ -805,6 +824,8 @@ struct kvm_vcpu_arch {
*/
bool enforce;
} pv_cpuid;
+
+ bool timer_passth_enable;
};
struct kvm_lpage_info {
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 47b8357b9751..38b8d80fa157 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -128,6 +128,11 @@ static bool __read_mostly enable_preemption_timer = 1;
module_param_named(preemption_timer, enable_preemption_timer, bool, S_IRUGO);
#endif
+static bool __read_mostly enable_timer_passth;
+#ifdef CONFIG_X86_64
+module_param_named(timer_passth, enable_timer_passth, bool, 0444);
+#endif
+
extern bool __read_mostly allow_smaller_maxphyaddr;
module_param(allow_smaller_maxphyaddr, bool, S_IRUGO);
@@ -220,6 +225,46 @@ static const struct {
[VMENTER_L1D_FLUSH_NOT_REQUIRED] = {"not required", false},
};
+#define TSC_DIVISOR 8
+static DEFINE_PER_CPU(struct timer_passth_info, passth_info);
+
+static int override_lapic_next_event(unsigned long delta,
+ struct clock_event_device *evt)
+{
+ struct timer_passth_info *local_timer_info;
+ u64 tsc;
+ u64 tscd;
+
+ local_timer_info = &per_cpu(passth_info, smp_processor_id());
+ tsc = rdtsc();
+ tscd = tsc + (((u64) delta) * TSC_DIVISOR);
+ local_timer_info->host_tscd = tscd;
+ wrmsrl(MSR_IA32_TSCDEADLINE, tscd);
+
+ return 0;
+}
+
+static void vmx_host_timer_passth_init(void *junk)
+{
+ struct timer_passth_info *local_timer_info;
+ int cpu = smp_processor_id();
+
+ local_timer_info = &per_cpu(passth_info, cpu);
+ local_timer_info->curr_dev = per_cpu(tick_cpu_device, cpu).evtdev;
+ local_timer_info->orig_set_next_event =
+ local_timer_info->curr_dev->set_next_event;
+ local_timer_info->curr_dev->set_next_event = override_lapic_next_event;
+}
+
+static void vmx_host_timer_restore(void *junk)
+{
+ struct timer_passth_info *local_timer_info;
+
+ local_timer_info = &per_cpu(passth_info, smp_processor_id());
+ local_timer_info->curr_dev->set_next_event =
+ local_timer_info->orig_set_next_event;
+}
+
#define L1D_CACHE_ORDER 4
static void *vmx_l1d_flush_pages;
@@ -7573,6 +7618,9 @@ static void vmx_migrate_timers(struct kvm_vcpu *vcpu)
static void hardware_unsetup(void)
{
+ if (enable_timer_passth)
+ on_each_cpu(vmx_host_timer_restore, NULL, 1);
+
if (nested)
nested_vmx_hardware_unsetup();
@@ -7884,6 +7932,9 @@ static __init int hardware_setup(void)
vmx_set_cpu_caps();
+ if (enable_timer_passth)
+ on_each_cpu(vmx_host_timer_passth_init, NULL, 1);
+
r = alloc_kvm_area();
if (r)
nested_vmx_hardware_unsetup();
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e545a8a613b1..5d353a9c9881 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -9911,6 +9911,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
vcpu->arch.pending_external_vector = -1;
vcpu->arch.preempted_in_kernel = false;
+ vcpu->arch.timer_passth_enable = false;
kvm_hv_vcpu_init(vcpu);
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 6c9c342dd0e5..bc50f4a1a7c0 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -26,6 +26,7 @@
* Tick devices
*/
DEFINE_PER_CPU(struct tick_device, tick_cpu_device);
+EXPORT_SYMBOL_GPL(tick_cpu_device);
/*
* Tick next event: keeps track of the tick time
*/
--
2.11.0
next prev parent reply other threads:[~2021-02-06 0:51 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-05 10:03 [RESEND RFC: timer passthrough 0/9] Support timer passthrough for VM Zhimin Feng
2021-02-05 10:03 ` Zhimin Feng [this message]
2021-02-05 12:28 ` [RFC: timer passthrough 1/9] KVM: vmx: hook set_next_event for getting the host tscd Peter Zijlstra
2021-02-05 18:11 ` Thomas Gleixner
2021-02-23 13:07 ` [External] " Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 2/9] KVM: vmx: enable host lapic timer offload preemtion timer Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 3/9] KVM: vmx: enable passthrough timer to guest Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 4/9] KVM: vmx: enable passth timer switch to sw timer Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 5/9] KVM: vmx: use tsc_adjust to enable tsc_offset timer passthrough Zhimin Feng
2021-02-05 12:43 ` Peter Zijlstra
2021-02-05 18:22 ` Thomas Gleixner
2021-02-05 10:03 ` [RFC: timer passthrough 6/9] KVM: vmx: check enable_timer_passth strictly Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 7/9] KVM: vmx: save the initial value of host tscd Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 8/9] KVM: vmx: Dynamically open or close the timer-passthrough for pre-vm Zhimin Feng
2021-02-05 10:03 ` [RFC: timer passthrough 9/9] KVM: vmx: query the state of timer-passth for vm Zhimin Feng
2021-02-05 13:34 ` [RESEND RFC: timer passthrough 0/9] Support timer passthrough for VM Paolo Bonzini
2021-02-08 18:13 ` Konrad Rzeszutek Wilk
2021-02-23 13:31 ` [External] " Zhimin Feng
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=20210205100317.24174-2-fengzhimin@bytedance.com \
--to=fengzhimin@bytedance.com \
--cc=bp@alien8.de \
--cc=fweisbec@gmail.com \
--cc=hpa@zytor.com \
--cc=jmattson@google.com \
--cc=joro@8bytes.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=seanjc@google.com \
--cc=tglx@linutronix.de \
--cc=vkuznets@redhat.com \
--cc=wanpengli@tencent.com \
--cc=x86@kernel.org \
--cc=zhanghaozhong@bytedance.com \
--cc=zhouyibo@bytedance.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).