Re: [PATCH] videobuf2-dma-sg: Support io userptr operations on io memory

[Marek Szyprowski <m.szyprowski@samsung.com>]

Hello,

On 2013-11-11 12:36, Matthias Wächter wrote:
> > @@ -180,7 +186,26 @@ static void *vb2_dma_sg_get_userptr(void
> > *alloc_ctx, unsigned long vaddr,
> >       if (!buf->pages)
> >               return NULL;
> >
> > -     num_pages_from_user = get_user_pages(current, current->mm,
> > +     buf->vma = find_vma(current->mm, vaddr);
> > +     if (!buf->vma) {
> > +             dprintk(1, "no vma for address %lu\n", vaddr);
> > +             return NULL;
> > +     }
> > +
> > +     if (vma_is_io(buf->vma)) {
> > +             for (num_pages_from_user = 0;
> > +                  num_pages_from_user < buf->num_pages;
> > +                  ++num_pages_from_user, vaddr += PAGE_SIZE) {
> > +                     unsigned long pfn;
> > +
> > +                     if (follow_pfn(buf->vma, vaddr, &pfn)) {
> > +                             dprintk(1, "no page for address %lu\n", vaddr);
> > +                             break;
> > +                     }
> > +                     buf->pages[num_pages_from_user] = pfn_to_page(pfn);
> > +             }
> > +     } else
> > +             num_pages_from_user = get_user_pages(current, current->mm,
> >                                            vaddr & PAGE_MASK,
> >                                            buf->num_pages,
> >                                            write,
>
> Can you safely assume that your userptr will cover only one vma? At least, get_user_pages (calling __get_user_pages) does not assume that and calls find_vma() whenever vma->vm_end is reached.

We care only about io mappings which cover only one vma. Such mappings
are created by other device drivers and can be used mainly for
zero-copy buffer sharing between multimedia devices. Although it is
technically possible to provide code for multiple vma, there will be
no real use case for it.

Best regards
-- 
Marek Szyprowski
Samsung R&D Institute Poland