[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20161102182206.GA7771@potion>
Date: Wed, 2 Nov 2016 19:22:07 +0100
From: Radim Krčmář <rkrcmar@...hat.com>
To: Paolo Bonzini <pbonzini@...hat.com>
Cc: linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
lcapitulino@...hat.com
Subject: Re: [PATCH] KVM: x86: drop TSC offsetting kvm_x86_ops to fix
KVM_GET/SET_CLOCK
2016-11-02 09:38+0100, Paolo Bonzini:
> Since commit a545ab6a0085 ("kvm: x86: add tsc_offset field to struct
> kvm_vcpu_arch", 2016-09-07) the offset between host and L1 TSC is
> cached and need not be fished out of the VMCS or VMCB. This means
> that we can implement adjust_tsc_offset_guest and read_l1_tsc
> entirely in generic code. The simplification is particularly
> significant for VMX code, where vmx->nested.vmcs01_tsc_offset
> was duplicating what is now in vcpu->arch.tsc_offset. Therefore
> the vmcs01_tsc_offset can be dropped completely.
>
> More importantly, this fixes KVM_GET_CLOCK/KVM_SET_CLOCK
> which, after commit 108b249c453d ("KVM: x86: introduce get_kvmclock_ns",
> 2016-09-01) called read_l1_tsc while the VMCS was not loaded.
> It thus returned bogus values on Intel CPUs.
>
> Fixes: 108b249c453dd7132599ab6dc7e435a7036c193f
> Reported-by: Roman Kagan <rkagan@...tuozzo.com>
> Signed-off-by: Paolo Bonzini <pbonzini@...hat.com>
> ---
> arch/x86/include/asm/kvm_host.h | 3 ---
> arch/x86/kvm/svm.c | 23 -----------------------
> arch/x86/kvm/vmx.c | 39 +++------------------------------------
> arch/x86/kvm/x86.c | 6 +++---
> 4 files changed, 6 insertions(+), 65 deletions(-)
Yay,
Reviewed-by: Radim Krčmář <rkrcmar@...hat.com>
Powered by blists - more mailing lists