New: tsc_offset is error for hot-plugged vcpu.

https://bugzilla.kernel.org/show_bug.cgi?id=115991

Bug ID: 115991
Summary: tsc_offset is error for hot-plugged vcpu.
Product: Virtualization
Version: unspecified
Kernel Version: 4.5
Hardware: x86-64
OS: Linux
Tree: Mainline
Status: NEW
Severity: normal
Priority: P1
Component: kvm
Assignee: [email protected]
Reporter: [email protected]
Regression: No

1. when hot plug vcpu, KVM will call kvm_write_tsc set the new vcpu's tsc_offset to current host tsc. The new vcpu's tsc_offsets is greatly different from existed vcpu's tsc_offset.

2. For redhat 6.x, the vm will show warning and turn off TSC clock.
kernel: CPU 1 got hotplugged
kernel: kvm-clock: cpu 1, msr 0:282367c1, secondary cpu clock
kernel: TSC synchronization [CPU#0 -> CPU#1]:
kernel: Measured 150076598781 cycles TSC warp between CPUs, turning off TSC clock.
kernel: Marking TSC unstable due to check_tsc_sync_source failed

3. I think should set the new vcpu's tsc_offset to kvm->arch.cur_tsc_offset if host tsc is stable and use_master_clock is true.

You are receiving this mail because:
You are watching the assignee of the bug.