On Sun, Jun 17, 2018 at 09:54:31PM +0000, Christopher Lameter wrote: > On Sat, 16 Jun 2018, john.hubbard@xxxxxxxxx wrote: > > > I've come up with what I claim is a simple, robust fix, but...I'm > > presuming to burn a struct page flag, and limit it to 64-bit arches, in > > order to get there. Given that the problem is old (Jason Gunthorpe noted > > that RDMA has been living with this problem since 2005), I think it's > > worth it. > > > > Leaving the new page flag set "nearly forever" is not great, but on the > > other hand, once the page is actually freed, the flag does get cleared. > > It seems like an acceptable tradeoff, given that we only get one bit > > (and are lucky to even have that). > > This is not robust. Multiple processes may register a page with the RDMA > subsystem. How do you decide when to clear the flag? I think you would > need an additional refcount for the number of times the page was > registered. And it's not just RDMA that is using get_user_pages. We have tons of users that do short, spurious get_user_pages do do zero copy operations. We can't leave the page in a wrecked state after that. > I still think the cleanest solution here is to require mmu notifier > callbacks and to not pin the page in the first place. If a NIC does not > support a hardware mmu then it can still simulate it in software by > holding off the ummapping the mmu notifier callback until any pending > operation is complete and then invalidate the mapping so that future > operations require a remapping (or refaulting). Sounds ok for RDMA, not going to help for most other users.
