CLOCK_MONOTONIC 和 CLOCK_MONOTONIC_RAW 的区别在于
CLOCK_MONOTONIC 的频率可以被调整，但是 CLOCK_MONOTONIC_RAW 完全不会: https://stackoverflow.com/questions/14270300/what-is-the-difference-between-clock-monotonic-clock-monotonic-raw
tk_data 是一个唯一全局变量，timekeeper 也是唯一的
/*
* The most important data for readout fits into a single 64 byte
* cache line.
*/
struct tk_data {
seqcount_raw_spinlock_t seq;
struct timekeeper timekeeper;
struct timekeeper shadow_timekeeper;
raw_spinlock_t lock;
} ____cacheline_aligned;
static struct tk_data tk_core;
masterclock 是每一个虚拟机一个的
在 kvm_arch 中:
struct kvm_arch {
// ...
u64 master_kernel_ns;
u64 master_cycle_now;
kvm->arch.kvmclock_offset 记录虚拟机开机的时候 kernel_ns 的 offset
MSR_KVM_SYSTEM_TIME 和 MSR_KVM_WALL_CLOCK
MSR_KVM_SYSTEM_TIME 中包含的是 pvti 的地址 MSR_KVM_WALL_CLOCK 处理墙上时间
Documentation/virt/kvm/x86/cpuid.rst
KVM_FEATURE_CLOCKSOURCE 0 kvmclock available at msrs
KVM_FEATURE_CLOCKSOURCE2 3 kvmclock available at msrs
在 arch/x86/include/uapi/asm/kvm_para.h 定义如下:
#define MSR_KVM_WALL_CLOCK 0x11
#define MSR_KVM_SYSTEM_TIME 0x12
/* Custom MSRs falls in the range 0x4b564d00-0x4b564dff */
#define MSR_KVM_WALL_CLOCK_NEW 0x4b564d00
#define MSR_KVM_SYSTEM_TIME_NEW 0x4b564d01
Documentation/virt/kvm/x86/msr.rst
MSR_KVM_WALL_CLOCK 的注册的位置为 kvm_register_clock
之所以有两个，细节可以看 boot_vcpu_runs_old_kvmclock 的 git blame
pvti 是如何沟通的
MSR_KVM_WALL_CLOCK_NEW 存储了 pvti 的地址
vmware 这个总结写的不错
https://www.cse.iitb.ac.in/~puru/courses/spring22/downloads/timekeeping.pdf
核心结构体
- struct clocksource
- struct clock_event_device
- struct tick_device
有趣的 trace clock
/sys/kernel/debug/tracing/README 中展示了 trace 的过程中也是可以使用多种时钟的:
trace_clock - change the clock used to order events
local: Per cpu clock but may not be synced across CPUs
global: Synced across CPUs but slows tracing down.
counter: Not a clock, but just an increment
uptime: Jiffy counter from time of boot
perf: Same clock that perf events use
x86-tsc: TSC cycle counter
这个整理的很好
四个组合都有，但是讨论他们没什么意义了
High-res Dynamic ticks | High-res Periodic ticks
-------------------------------------------------
Low-res Dynamic ticks | Low-res Periodic ticks
windows 在 kvm 环境中运行，也是使用的 kvmclock 吧?
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index 2f66e63b880a..726837c79955 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -4718,6 +4718,7 @@ static int kvm_get_msrs(X86CPU *cpu)
break;
case MSR_KVM_SYSTEM_TIME:
env->system_time_msr = msrs[i].data;
+ printf("[martins3:%s:%d] %lx\n", __FUNCTION__, __LINE__, env->system_time_msr);
break;
case MSR_KVM_WALL_CLOCK:
env->wall_clock_msr = msrs[i].data;
启动之后可以获取到如下日志:
[martins3:kvm_get_msrs:4721] e8601
不过这个似乎是 seabios 来初始化的，更多的证据是从外部直接修改 pvti ，看看 windows 的时间会不会跳变。
配置 timer 的方法
通过 MSR_IA32_TSC_DEADLINE
@[
lapic_next_deadline+5
clockevents_program_event+138
hrtimer_interrupt+291
__sysvec_apic_timer_interrupt+85
sysvec_apic_timer_interrupt+110
asm_sysvec_apic_timer_interrupt+26
cpuidle_enter_state+220
cpuidle_enter+45
do_idle+462
cpu_startup_entry+41
start_secondary+291
common_startup_64+318
]: 3584
获取时间的经典路径
- sysvec_apic_timer_interrupt
- irq_enter_rcu
- tick_irq_enter
- tick_nohz_irq_enter
- ktime_get
- timekeeping_get_ns
- timekeeping_get_delta
- tk_clock_read
- kvm_clock_get_cycles
- tk_clock_read
- timekeeping_get_delta
- timekeeping_get_ns
- ktime_get
- tick_nohz_irq_enter
- tick_irq_enter
- irq_enter_rcu
继续看看
- 为什么 qemu 中也存在 kvm-clock 的支持
- https://zhuanlan.zhihu.com/p/665543594
- https://tcbbd.moe/linux/qemu-kvm/kvm-time/#more
- https://github.com/GiantVM/KVM-Annotation/wiki/Kvmclock
- https://rwmj.wordpress.com/2010/10/15/kvm-pvclock/
- https://lkml.org/lkml/2010/4/15/355
- https://github.com/WCharacter/RDTSC-KVM-Handler
See “Changes to Instruction Behavior in VMX Non-Root Operation” in Chapter 25 of the Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 3C, for more information about the behavior of this instruction in VMX non-root operation.
- https://www.redhat.com/en/blog/avoiding-clock-drift-vms
本站所有文章转发 CSDN 将按侵权追究法律责任，其它情况随意。