Rusty Russell wrote:
4) Access to the native versions is trivial for KVM, lguest, Xen and
others who might want it.
kvm will be quite happy with it, as long as the x86_64 version has a
similar API.
include/asm-x86 anyone? ;-)
--
Do not meddle in the internals of kernels, for they are
On Thu, 2007-03-22 at 10:09 +0200, Avi Kivity wrote:
> Right, scratch that. Was confused by rdmsr_safe().
Heh... me too 8)
> > The behaviour change (don't oops when an invalid rdmsr is used) was
> > there with CONFIG_PARAVIRT=y, the cleanup just made !CONFIG_PARAVIRT the
> > same. Is it importa
Rusty Russell wrote:
> The behaviour change (don't oops when an invalid rdmsr is used) was
> there with CONFIG_PARAVIRT=y, the cleanup just made !CONFIG_PARAVIRT the
> same. Is it important?
>
We could always fling a BUG_ON in the non-safe versions, which would
probably be more informative tha
Rusty Russell wrote:
+#define rdmsr(msr,val1,val2) \
+ do {\
+ int __err; \
+ unsigned long long __val = native_re
On Thu, 2007-03-22 at 09:30 +0200, Avi Kivity wrote:
> Rusty Russell wrote:
> > +static inline unsigned long long native_read_msr(unsigned int msr, int
> > *err)
> > +{
> > + unsigned long long val;
> > +
> > + asm volatile("2: rdmsr ; xorl %0,%0\n"
> > +"1:\n\t"
> > +
Rusty Russell wrote:
+static inline unsigned long long native_read_msr(unsigned int msr, int *err)
+{
+ unsigned long long val;
+
+ asm volatile("2: rdmsr ; xorl %0,%0\n"
+"1:\n\t"
+".section .fixup,\"ax\"\n\t"
+"3: movl %3,
6 matches
Mail list logo
