On Wed, Nov 06, 2013 at 06:50:05PM +0100, Paolo Bonzini wrote: > Il 06/11/2013 18:48, Michael S. Tsirkin ha scritto: > > This just makes the symproms go away. > > That's correct. > > > The real bug is exec ignores high address bits during page > > lookups. It should fail on invalid address not access > > a random page. > > I'll send a patch. > > The real real bug is that all address spaces should be 2^64, which you > said you consider too intrusive a patch.
Because this will affect performance in unpredicatable way. We can't make such changes in 1.7 IMHO: it would need much more than just a quick "works for me". > I don't feel confident > changing phys_page_find, even if it's just 2 lines. > > Paolo Well it's *obviously* broken if address is outside target address space. Take a look at the patch first, then argue. -- MST