Re: [PATCH 4/9] Vmi fix highpte

[Jeremy Fitzhardinge <jeremy@goop.org>]

Zachary Amsden wrote:
> I've been sending out this particular patch or a variant of it for a
> long time.  It did get lost for a while during the paravirt-ops
> conversion, however.  You're the first to give any feedback on it.

Oops, I guess so.  I've been doing a lot more Xen pagetable work for the
last couple of weeks, so it caught my eye this time.

>>>   Do you have any objection to me creating a patch for -mm tree that
>>> implements kmap_atomic_pte the way you have described above and
>>> attaching it to the Xen patch series, but leaving the current patch as
>>> is for now?
>>>     
>>
>> Not particularly, but it seems odd to put something in knowing its going
>> to be immediately replaced.  What's the urgency?
>>   
>
> Better to keep what is known working for now, even if it is going to
> be replaced later... code is easy to change in development cycles,
> less easy to fix when nearing release.  It really is easy to mess up
> one of the pte conversions by, say, shift the wrong value or calculate
> wrong or PAE dependent PTE offset.

Those are bugs that can occur, but they don't apply in this case.  The
vmi implementation of kmap_atomic_pte() would be:

static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type)
{
	void *ptep = kmap_atomic(page, type);
	vmi_map_pt_hook(type, ptep, page_to_pfn(page));
	return ptep;
}

Right?  Which is functionally identical to the code in your patch,
except wrapped up in a new function.

    J