]> git.proxmox.com Git - mirror_ubuntu-kernels.git/commit
kvmclock: fix TSC calibration for nested guests
authorPeng Hao <peng.hao2@zte.com.cn>
Sat, 14 Jul 2018 15:28:29 +0000 (23:28 +0800)
committerPaolo Bonzini <pbonzini@redhat.com>
Wed, 18 Jul 2018 09:43:17 +0000 (11:43 +0200)
commite10f7805032365cc11c739a97f226ebb48aee042
treefe7cf33daa3ef5a8fc7ae2a1ae64a417686c9860
parent2307af1c4b2e0ad886f30e31739845322cbd328b
kvmclock: fix TSC calibration for nested guests

Inside a nested guest, access to hardware can be slow enough that
tsc_read_refs always return ULLONG_MAX, causing tsc_refine_calibration_work
to be called periodically and the nested guest to spend a lot of time
reading the ACPI timer.

However, if the TSC frequency is available from the pvclock page,
we can just set X86_FEATURE_TSC_KNOWN_FREQ and avoid the recalibration.
'refine' operation.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Peng Hao <peng.hao2@zte.com.cn>
[Commit message rewritten. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/kernel/kvmclock.c