On 01/11/2018 10:05 AM, Willy Tarreau wrote: >> That's exactly what it does for now, but Linus doesn't like it at all. >> So I'll switch it back to per-mm + per-CPU variable. Well he has a valid >> point regarding the pgd and _PAGE_NX setting. point Now at least we know >> the change is minimal if we have a good reason for doing differently >> later. > Yuck, I hate this. Are you really going to wire it up complete with > all the IPIs needed to get the thing synced up right?
Well, on the bright side, we don't need IPIs when _removing_ NX. We can just handle those like a spurious fault. But, when re-enabling it, we need all the TLB flushing for all the CPUs that have run with the un-NX'd page tables. I guess that's the one benefit if you come up with an API that only allows "disable PTI" while a task is running but leaves execve()/fork() as places that implicitly reenable PTI.
