On Tue Apr 29, 2025 at 9:28 AM BST, Jan Beulich wrote: > On 28.04.2025 21:57, Ariadne Conill wrote: >> Previously Xen placed the hypercall page at the highest possible MFN, >> but this caused problems on systems where there is more than 36 bits >> of physical address space. > > Hmm, I should have asked already on the earlier version: What kinds of > problems are these, beyond ... > >> In general, it also seems unreliable to assume that the highest possible >> MFN is not already reserved for some other purpose. > > ... this particular aspect? I find it puzzling that such problems would > depend on the number of physical address bits. > > Jan
Pagefault on access (due to reserved bits being set) on access to the hypercall page. The available guest-physical address space doesn't seem to be as wide as advertised, though I didn't carry enough tests to single this as the only explanation. Seeing how we don't really know what's already on the last mfn this seems like a strict improvement irrespective of the actual cause of the fault. Cheers, Alejandro
