Re: [PATCH 1/1] direct mmio for passthrough - kernel part

[Avi Kivity <avi@qumranet.com>]

Anthony Liguori wrote:
>
> You could get away with supporting reserved RAM another way though.  
> If we refactored the MMU to use hfn_t instead of struct page, you 
> would then need a mechanism to mmap() reserved ram into userspace 
> (similar to ioremap I guess).  In fact, you may be able to just lie 
> and claim reserved ram is IO ram.

We'd need it to return a (hfn_t, page_type_t) pair so the caller would 
know whether to use page refcounting or not.  The caller might be able 
to ask Linux whether the page has a struct page or not, but it gets hairy.

That's why I'm interested in the possibility of adding struct pages at 
runtime. vmemmap is needed for many other reasons (NUMA, memory hotplug, 
sparse SGI memory) so it's reasonable that it will be enabled on most 
distros.

-- 
Any sufficiently difficult bug is indistinguishable from a feature.


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace