Re: [PATCH 1/5] mm: Introduce mm_struct.has_pinned

From: Jason Gunthorpe
Date: Thu Sep 24 2020 - 12:51:56 EST


On Thu, Sep 24, 2020 at 10:35:17AM -0400, Peter Xu wrote:

> If so, I'd hope you won't disagree that I still move the get_page() out of the
> "if (wp)". Not only it's a shared operation no matter whether "if (wp)" or
> not, but I'm afraid it would confuse future readers on a special ordering on
> the get_page() and the wrprotect(), especially with the comment above.

Sure, you could add a comment before the page_maybe_dma_pinned that it
could be fused with get_page()

> Yes. It's kind of related here on whether we can still use wrprotect() to
> guard against fast-gup, though. So my understanding is that we should still at
> least need the other patch [1] that I proposed in the other thread to force
> break-cow for read-only gups (that patch is not only for fast-gup, of course).

Probably, I haven't intensively studied that patch, and it should go
along with edits to some of the callers..

> But I agree that should be another bigger topic. I hope we don't need to pick
> that patch up someday by another dma report on read-only pinned pages...

In RDMA we found long ago that read only pins don't work well, I think
most other places are likely the same - the problems are easy enough
to hit. Something like your COW break patch on read is really needed
to allow read-only GUP.

> Regarding the solution here, I think we can also cover read-only fast-gup too
> in the future - IIUC what we need to do is to make it pte_protnone() instead of
> pte_wrprotect(), then in the fault handler we should identify this special
> pte_protnone() against numa balancing (change_prot_numa()). I think it should
> work fine too, iiuc, because I don't think we should migrate a page at all if
> it's pinned for any reason...

With your COW breaking patch the read only fast-gup should break the
COW because of the write protect, just like for the write side. Not
seeing why we need to do something more?

Jason