From: Gleb Natapov <gleb@redhat.com>
To: Marcelo Tosatti <mtosatti@redhat.com>
Cc: kvm@vger.kernel.org, Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: VMX: x86: handle host TSC calibration failure
Date: Tue, 12 Mar 2013 13:18:52 +0200 [thread overview]
Message-ID: <20130312111851.GW11223@redhat.com> (raw)
In-Reply-To: <20130312021024.GA21328@amt.cnet>
On Mon, Mar 11, 2013 at 11:10:24PM -0300, Marcelo Tosatti wrote:
>
> If the host TSC calibration fails, tsc_khz is zero (see tsc_init.c).
> Handle such case properly in KVM (instead of dividing by zero).
>
> https://bugzilla.redhat.com/show_bug.cgi?id=859282
>
> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
>
Fixed small space/tab problem and applied. Thanks.
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 35b4912..19741b9 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -1079,6 +1079,10 @@ static void kvm_set_tsc_khz(struct kvm_vcpu *vcpu, u32 this_tsc_khz)
> u32 thresh_lo, thresh_hi;
> int use_scaling = 0;
>
> + /* tsc_khz can be zero if TSC calibration fails */
> + if (this_tsc_khz == 0)
> + return;
> +
> /* Compute a scale to convert nanoseconds in TSC cycles */
> kvm_get_time_scale(this_tsc_khz, NSEC_PER_SEC / 1000,
> &vcpu->arch.virtual_tsc_shift,
> @@ -1156,20 +1160,23 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, struct msr_data *msr)
> ns = get_kernel_ns();
> elapsed = ns - kvm->arch.last_tsc_nsec;
>
> - /* n.b - signed multiplication and division required */
> - usdiff = data - kvm->arch.last_tsc_write;
> + if (vcpu->arch.virtual_tsc_khz) {
> + /* n.b - signed multiplication and division required */
> + usdiff = data - kvm->arch.last_tsc_write;
> #ifdef CONFIG_X86_64
> - usdiff = (usdiff * 1000) / vcpu->arch.virtual_tsc_khz;
> + usdiff = (usdiff * 1000) / vcpu->arch.virtual_tsc_khz;
> #else
> - /* do_div() only does unsigned */
> - asm("idivl %2; xor %%edx, %%edx"
> - : "=A"(usdiff)
> - : "A"(usdiff * 1000), "rm"(vcpu->arch.virtual_tsc_khz));
> + /* do_div() only does unsigned */
> + asm("idivl %2; xor %%edx, %%edx"
> + : "=A"(usdiff)
> + : "A"(usdiff * 1000), "rm"(vcpu->arch.virtual_tsc_khz));
> #endif
> - do_div(elapsed, 1000);
> - usdiff -= elapsed;
> - if (usdiff < 0)
> - usdiff = -usdiff;
> + do_div(elapsed, 1000);
> + usdiff -= elapsed;
> + if (usdiff < 0)
> + usdiff = -usdiff;
> + } else
> + usdiff = USEC_PER_SEC; /* disable TSC match window below */
>
> /*
> * Special case: TSC write with a small delta (1 second) of virtual
--
Gleb.
prev parent reply other threads:[~2013-03-12 11:18 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-12 2:10 VMX: x86: handle host TSC calibration failure Marcelo Tosatti
2013-03-12 11:18 ` Gleb Natapov [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130312111851.GW11223@redhat.com \
--to=gleb@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=mtosatti@redhat.com \
--cc=pbonzini@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox