[PATCH] KVM: X86: Fix softlockup when get the current kvmclock timestamp

From: Wanpeng Li
Date: Sun Nov 05 2017 - 19:55:34 EST


From: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>

watchdog: BUG: soft lockup - CPU#6 stuck for 22s! [qemu-system-x86:10185]
CPU: 6 PID: 10185 Comm: qemu-system-x86 Tainted: G OE 4.14.0-rc4+ #4
RIP: 0010:kvm_get_time_scale+0x4e/0xa0 [kvm]
Call Trace:
? get_kvmclock_ns+0xa3/0x140 [kvm]
get_time_ref_counter+0x5a/0x80 [kvm]
kvm_hv_process_stimers+0x120/0x5f0 [kvm]
? kvm_hv_process_stimers+0x120/0x5f0 [kvm]
? preempt_schedule+0x27/0x30
? ___preempt_schedule+0x16/0x18
kvm_arch_vcpu_ioctl_run+0x4b4/0x1690 [kvm]
? kvm_arch_vcpu_load+0x47/0x230 [kvm]
kvm_vcpu_ioctl+0x33a/0x620 [kvm]
? kvm_vcpu_ioctl+0x33a/0x620 [kvm]
? kvm_vm_ioctl_check_extension_generic+0x3b/0x40 [kvm]
? kvm_dev_ioctl+0x279/0x6c0 [kvm]
do_vfs_ioctl+0xa1/0x5d0
? __fget+0x73/0xa0
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x1e/0xa9

This can be reproduced when running kvm-unit-tests/hyperv_stimer.flat and
cpu-hotplug stress simultaneously. kvm_get_time_scale() takes too long which
results in softlockup.

This patch fixes it by disabling preemption just for rdtsc() and __this_cpu_read()
section in order to reduce preemption disable time, the per-CPU variable is just
read operation and the master clock is active, so it is as safe as how it is used
in kvm_guest_time_update().

Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Cc: Radim KrÄmÃÅ <rkrcmar@xxxxxxxxxx>
Signed-off-by: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>
---
arch/x86/kvm/x86.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 34c85aa..2542f9b 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1781,6 +1781,7 @@ u64 get_kvmclock_ns(struct kvm *kvm)
struct kvm_arch *ka = &kvm->arch;
struct pvclock_vcpu_time_info hv_clock;
u64 ret;
+ unsigned long this_tsc_khz, host_tsc;

spin_lock(&ka->pvclock_gtod_sync_lock);
if (!ka->use_master_clock) {
@@ -1795,13 +1796,17 @@ u64 get_kvmclock_ns(struct kvm *kvm)
/* both __this_cpu_read() and rdtsc() should be on the same cpu */
get_cpu();

- kvm_get_time_scale(NSEC_PER_SEC, __this_cpu_read(cpu_tsc_khz) * 1000LL,
- &hv_clock.tsc_shift,
- &hv_clock.tsc_to_system_mul);
- ret = __pvclock_read_cycles(&hv_clock, rdtsc());
+ this_tsc_khz = __this_cpu_read(cpu_tsc_khz);
+ host_tsc = rdtsc();

put_cpu();

+ /* With all the info we got, fill in the values */
+ kvm_get_time_scale(NSEC_PER_SEC, this_tsc_khz * 1000LL,
+ &hv_clock.tsc_shift,
+ &hv_clock.tsc_to_system_mul);
+ ret = __pvclock_read_cycles(&hv_clock, host_tsc);
+
return ret;
}

--
2.7.4