On Thu, Apr 08, 2021 at 08:13:43AM +0100, Matthew Wilcox wrote: > On Thu, Apr 08, 2021 at 09:00:26AM +0200, Peter Zijlstra wrote: > > On Wed, Apr 07, 2021 at 10:27:12PM +0100, Matthew Wilcox wrote: > > > Doing I/O without any lock held already works; it just uses the file > > > refcount. It would be better to use a vma refcount, as I already said. > > > > The original workload that I developed SPF for (waaaay back when) was > > prefaulting a single huge vma. Using a vma refcount was a total loss > > because it resulted in the same cacheline contention that down_read() > > was having. > > > > As such, I'm always incredibly sad to see mention of vma refcounts. > > They're fundamentally not solving the problem :/ > > OK, let me outline my locking scheme because I think it's rather better > than Michel's. The vma refcount is the slow path. > > 1. take the RCU read lock > 2. walk the pgd/p4d/pud/pmd > 3. allocate page tables if necessary. *handwave GFP flags*. The problem with allocating page-tables was that you can race with zap_page_range() if you're not holding mmap_sem, and as such can install a page-table after, in which case it leaks. IIRC that was solvable, but it did need a bit of care. > 4. walk the vma tree > 5. call ->map_pages I can't remember ->map_pages().. I think that's 'new'. git-blame tells me that's 2014, and I did the original SPF in 2010. Yes, that looks like a useful thing to have, it does the non-blocking part of ->fault(). I suppose the thing missing here is that if ->map_pages() does not return a page, we have: goto 9 > 6. take ptlock > 7. insert page(s) > 8. drop ptlock > if this all worked out, we're done, drop the RCU read lock and return. > 9. increment vma refcount > 10. drop RCU read lock > 11. call ->fault > 12. decrement vma refcount And here we do 6-8 again, right? > Compared to today, where we bump the refcount on the file underlying the > vma, this is _better_ scalability -- different mappings of the same file > will not contend on the file's refcount. > > I suspect your huge VMA was anon, and that wouldn't need a vma refcount > as faulting in new pages doesn't need to do I/O, just drop the RCU > lock, allocate and retry. IIRC yes, it was either a huge matrix setup or some database thing, I can't remember. But the thing was, we didn't have that ->map_pages(), so we had to call ->fault(), which can sleep, so I had to use SRCU across the whole thing (or rather, I hacked up preemptible-rcu, because SRCU was super primitive back then). It did kick start significant SRCU rework IIRC. Anyway, that's all ancient history.
