Re: [PATCH v3 11/11] KVM: x86: Introduce KVM_SET_PAGE_ENC_BITMAP ioctl

[Steve Rutherford]

On Thu, Nov 14, 2019 at 5:22 PM Steve Rutherford <srutherford@xxxxxxxxxx> wrote:
>
> On Wed, Jul 10, 2019 at 1:13 PM Singh, Brijesh <brijesh.singh@xxxxxxx> wrote:
> >
> >  struct kvm_arch_async_pf {
> > diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
> > index e675fd89bb9a..31653e8d5927 100644
> > --- a/arch/x86/kvm/svm.c
> > +++ b/arch/x86/kvm/svm.c
> > @@ -7466,6 +7466,47 @@ static int svm_get_page_enc_bitmap(struct kvm *kvm,
> >         return ret;
> >  }
> >
> > +static int svm_set_page_enc_bitmap(struct kvm *kvm,
> > +                                  struct kvm_page_enc_bitmap *bmap)
> > +{
> > +       struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info;
> > +       unsigned long gfn_start, gfn_end;
> > +       unsigned long *bitmap;
> > +       unsigned long sz, i;
> > +       int ret;
> > +
> > +       if (!sev_guest(kvm))
> > +               return -ENOTTY;
> > +
> > +       gfn_start = bmap->start_gfn;
> > +       gfn_end = gfn_start + bmap->num_pages;
> > +
> > +       sz = ALIGN(bmap->num_pages, BITS_PER_LONG) / 8;
> > +       bitmap = kmalloc(sz, GFP_KERNEL);
>
> This kmalloc should probably be either a vmalloc or kvmalloc. The max
> size, if I'm reading kmalloc correctly, is 2^10 pages. That's 4MB,
> which should correspond to a bitmap for a 128GB VM, which is a
> plausible VM size.
>
> --Steve

Ignore this, you are clearly never going to copy that much in one
call, and the backing bitmap correctly uses kvmalloc.