Re: [PATCH RFC] uaccess: user_access_begin_after_access_ok()

[Michael S. Tsirkin]

On Wed, Jun 03, 2020 at 02:36:46PM +0800, Jason Wang wrote:
> 
> On 2020/6/3 äå2:30, Michael S. Tsirkin wrote:
> > On Wed, Jun 03, 2020 at 02:23:08PM +0800, Jason Wang wrote:
> > > > BTW now I re-read it I don't understand __vhost_get_user_slow:
> > > > 
> > > > 
> > > > static void __user *__vhost_get_user_slow(struct vhost_virtqueue *vq,
> > > >                                             void __user *addr, unsigned int size,
> > > >                                             int type)
> > > > {
> > > >           int ret;
> > > > 
> > > >           ret = translate_desc(vq, (u64)(uintptr_t)addr, size, vq->iotlb_iov,
> > > >                                ARRAY_SIZE(vq->iotlb_iov),
> > > >                                VHOST_ACCESS_RO);
> > > > 
> > > > ..
> > > > }
> > > > 
> > > > how does this work? how can we cast a pointer to guest address without
> > > > adding any offsets?
> > > 
> > > I'm not sure I get you here. What kind of offset did you mean?
> > > 
> > > Thanks
> > OK so points:
> > 
> > 1. type argument seems unused. Right?
> 
> 
> Yes, we can remove that.
> 
> 
> > 2. Second argument to translate_desc is a GPA, isn't it?
> 
> 
> No, it's IOVA, this function will be called only when IOTLB is enabled.
> 
> Thanks

Right IOVA. Point stands how does it make sense to cast
a userspace pointer to an IOVA? I guess it's just
because it's talking to qemu actually, so it's abusing
the notation a bit ...

> 
> >     Here we cast a userspace address to this type. What if it
> >     matches a valid GPA by mistake?
> >