Prev: genirq: reliably replay pending edge-triggered irq
Next: Crypto Fixes for 2.6.34
From: Marcelo Tosatti on 27 Apr 2010 09:50
On Mon, Apr 26, 2010 at 01:46:28PM -0400, Glauber Costa wrote:
> If the HV told us we can fully trust the TSC, skip any
> correction
>
> Signed-off-by: Glauber Costa <glommer(a)redhat.com>
> ---
> arch/x86/include/asm/kvm_para.h | 5 +++++
> arch/x86/include/asm/pvclock-abi.h | 1 +
> arch/x86/kernel/kvmclock.c | 3 +++
> arch/x86/kernel/pvclock.c | 4 ++++
> 4 files changed, 13 insertions(+), 0 deletions(-)
>
> diff --git a/arch/x86/include/asm/kvm_para.h b/arch/x86/include/asm/kvm_para.h
> index f019f8c..615ebb1 100644
> --- a/arch/x86/include/asm/kvm_para.h
> +++ b/arch/x86/include/asm/kvm_para.h
> @@ -21,6 +21,11 @@
> */
> #define KVM_FEATURE_CLOCKSOURCE2 3
>
> +/* The last 8 bits are used to indicate how to interpret the flags field
> + * in pvclock structure. If no bits are set, all flags are ignored.
> + */
> +#define KVM_FEATURE_CLOCKSOURCE_STABLE_TSC 0xf8
> +
> #define MSR_KVM_WALL_CLOCK 0x11
> #define MSR_KVM_SYSTEM_TIME 0x12

Please drop this. Its not certain what is the best method to reduce
contention on the global variable. Whatever it is, can be done later.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Glauber Costa on 27 Apr 2010 11:20
On Tue, Apr 27, 2010 at 10:40:13AM -0300, Marcelo Tosatti wrote:
> On Mon, Apr 26, 2010 at 01:46:28PM -0400, Glauber Costa wrote:
> > If the HV told us we can fully trust the TSC, skip any
> > correction
> >
> > Signed-off-by: Glauber Costa <glommer(a)redhat.com>
> > ---
> > arch/x86/include/asm/kvm_para.h | 5 +++++
> > arch/x86/include/asm/pvclock-abi.h | 1 +
> > arch/x86/kernel/kvmclock.c | 3 +++
> > arch/x86/kernel/pvclock.c | 4 ++++
> > 4 files changed, 13 insertions(+), 0 deletions(-)
> >
> > diff --git a/arch/x86/include/asm/kvm_para.h b/arch/x86/include/asm/kvm_para.h
> > index f019f8c..615ebb1 100644
> > --- a/arch/x86/include/asm/kvm_para.h
> > +++ b/arch/x86/include/asm/kvm_para.h
> > @@ -21,6 +21,11 @@
> > */
> > #define KVM_FEATURE_CLOCKSOURCE2 3
> >
> > +/* The last 8 bits are used to indicate how to interpret the flags field
> > + * in pvclock structure. If no bits are set, all flags are ignored.
> > + */
> > +#define KVM_FEATURE_CLOCKSOURCE_STABLE_TSC 0xf8
> > +
> > #define MSR_KVM_WALL_CLOCK 0x11
> > #define MSR_KVM_SYSTEM_TIME 0x12
>
> Please drop this. Its not certain what is the best method to reduce
> contention on the global variable. Whatever it is, can be done later.
You tell me to drop it, avi tells me to write this, I am happy to pick
any one of the two options. Just don't want to get ping-ponged between
haveit-dropit.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Glauber Costa on 27 Apr 2010 15:00
On Tue, Apr 27, 2010 at 09:03:55PM +0300, Avi Kivity wrote:
> On 04/27/2010 04:40 PM, Marcelo Tosatti wrote:
> >On Mon, Apr 26, 2010 at 01:46:28PM -0400, Glauber Costa wrote:
> >>If the HV told us we can fully trust the TSC, skip any
> >>correction
> >>
> >Please drop this. Its not certain what is the best method to reduce
> >contention on the global variable. Whatever it is, can be done later.
> >
>
> The problem is lead time. If we (or the cpu vendors) fix this in
> the hypervisor/processors in a year, then we want existing guests to
> take advantage of it immediately.
100 % Agreed.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: genirq: reliably replay pending edge-triggered irq
Next: Crypto Fixes for 2.6.34