Re: User switchable HW mappings & cie

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Mon, 2006-10-09 at 20:36 +0200, Ingo Oeser wrote:
> Hi all,
> 
> On Monday, 9. October 2006 13:51, Benjamin Herrenschmidt wrote:
> > > One problem that occurs is that the rule for ptes with non-backing 
> > > struct pages
> > > Which I think was introduced in 2.6.16:
> > > 
> > >     pfn_of_page == vma->vm_pgoff + ((addr - vma->vm_start) >> PAGE_SHIFT)
> > > 
> > > cannot be honored, at least not with the DRM memory manager, since the 
> > > graphics object will be associated with a vma and not the underlying 
> > > physical address. User space will have vma->vm_pgoff as a handle to the 
> > > object, which may move around in graphics memory.
> > 
> > That's a problem with VM_PFNMAP set indeed. get_user_pages() is a
> > non-issue with VM_IO set too but I'm not sure about other code path that
> > might try to hit here... though I think we don't hit that if MAP_SHARED,
> > Nick ?
> 
> Istn't this just a non-linear PFN mapping, you are describing here?
> 
> Nick: 
> 	Cant your new fault consolidation code handle that?
> 	AFAICS your new .fault handler just gets the
> 	vma and pgoff and install the matching PTE via install_THINGIE()
> 	or vm_insert_THINGIE()
> 
> Or do I miss sth. here?

It is somewhat yes.

Ben.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/