[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] [PATCH 21/34] x86/mm: p2m_flush and nvcpu_flush are HVM only
On Wed, Aug 22, 2018 at 03:14:12PM +0100, Wei Liu wrote: > On Tue, Aug 21, 2018 at 02:01:12AM -0600, Jan Beulich wrote: > > >>> On 17.08.18 at 17:12, <wei.liu2@xxxxxxxxxx> wrote: > > > p2m_flush is only called by HAP code, nvcpu_flush is only useful for > > > nestedhvm, both of which depend on HVM support. > > > > > > Enclose their code in CONFIG_HVM. Add assertions. > > > > But why do you put the #ifdef-s inside the functions, rather than > > around them? From what you say, without CONFIG_HVM no caller > > should exist. > > > > Jan > > > > > Signed-off-by: Wei Liu <wei.liu2@xxxxxxxxxx> > > > --- > > > xen/arch/x86/mm/p2m.c | 6 ++++++ > > > 1 file changed, 6 insertions(+) > > > > > > diff --git a/xen/arch/x86/mm/p2m.c b/xen/arch/x86/mm/p2m.c > > > index 1089b86..1a04b34 100644 > > > --- a/xen/arch/x86/mm/p2m.c > > > +++ b/xen/arch/x86/mm/p2m.c > > > @@ -1785,10 +1785,13 @@ p2m_flush_table(struct p2m_domain *p2m) > > > void > > > p2m_flush(struct vcpu *v, struct p2m_domain *p2m) > > This is called by HAP code which I didn't put under CONFIG_HVM yet. > > I will see about reworking that part to put HAP under CONFIG_HVM. > > > > { > > > +#if CONFIG_HVM > > > + ASSERT(is_hvm_vcpu(v)); > > > ASSERT(v->domain == p2m->domain); > > > vcpu_nestedhvm(v).nv_p2m = NULL; > > > p2m_flush_table(p2m); > > > hvm_asid_flush_vcpu(v); > > > +#endif > > > } > > > > > > void > > > @@ -1839,8 +1842,11 @@ static void assign_np2m(struct vcpu *v, struct > > > p2m_domain *p2m) > > > > > > static void nvcpu_flush(struct vcpu *v) > > This is referenced by another HVM function. Maybe I can reshuffle code a > bit more to make it work. It appears that more code can be put under CONFIG_HVM. I have done that. Wei. _______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxxxxxxxxx https://lists.xenproject.org/mailman/listinfo/xen-devel
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |