Re: [Linux PATCH] Fix to hugepages to work around new PWT handling

[Jeremy Fitzhardinge <jeremy@goop.org>]

On 06/09/2010 08:26 PM, Dave McCracken wrote:
>> But in general kernel code shouldn't be just nakedly setting present on
>> the pte without also remaking the whole thing.  That doesn't happen with
>> normal ptes, and it probably shouldn't happen with huge ptes.  Forcing
>> present on a pte at this level seems very bogus.   Why not change the
>> upper code to set present if that's want it wants?
>>
>> I'll skip this chunk for now.
>>     
> Um, this is the upper level code.  The entire purpose of make_huge_pte is to 
> construct a present huge pte from page and pgprot. The problem is that the 
> original code makes the pte, then sets the present bit via pte_mkhuge().  This 
> means the Xen-specific macro that triggers on present is misled and doesn't do 
> the pfn_to_mfn().  Without this patch hugepages is handing pfns to the 
> hypervisor to map instead of mfns.
>   

In principle, setting present should cause the pte to be converted from
pfn to mfn, but I don't think that ever happens with normal ptes (since
non-present ptes contain swap info).  But I don't see where a huge pte
gets present set; pte_mkhuge itself doesn't do anything except set PSE.

    J