On 15.04.2020 10:21, Jan Beulich wrote: > While looking at "x86_64/mm: map and unmap page tables in > destroy_compat_m2p_mapping" it occurred to me that the mappings > changed there can be dropped altogether, as can the linear > address range used for this. Note that both patches have only > been lightly tested so far, I guess in particular the 2nd one > may still have issues. > > 1: x86: drop unnecessary page table walking in compat r/o M2P handling > 2: x86: drop high compat r/o M2P table address range
Just fyi - with the reviews I've got I'm intending to commit this later this week, unless I hear otherwise soon. Jan
