Re: [RFC PATCH v1 1/6] fs/proc/task_mmu: Fix pte update and tlb maintenance ordering in pagemap_scan_pmd_entry()

[Jann Horn <jannh@xxxxxxxxxx>]

On Fri, May 30, 2025 at 6:45 PM Ryan Roberts <ryan.roberts@xxxxxxx> wrote:
> On 30/05/2025 17:26, Jann Horn wrote:
> > On Fri, May 30, 2025 at 4:04 PM Ryan Roberts <ryan.roberts@xxxxxxx> wrote:
> >> pagemap_scan_pmd_entry() was previously modifying ptes while in lazy mmu
> >> mode, then performing tlb maintenance for the modified ptes, then
> >> leaving lazy mmu mode. But any pte modifications during lazy mmu mode
> >> may be deferred until arch_leave_lazy_mmu_mode(), inverting the required
> >> ordering between pte modificaiton and tlb maintenance.
> >>
> >> Let's fix that by leaving mmu mode, forcing all the pte updates to be
> >> actioned, before doing the tlb maintenance.
> >>
> >> This is a theorectical bug discovered during code review.
> >>
> >> Fixes: 52526ca7fdb9 ("fs/proc/task_mmu: implement IOCTL to get and 
> >> optionally clear info about PTEs")
> >
> > Hmm... isn't lazy mmu mode supposed to also delay TLB flushes, and
> > preserve the ordering of PTE modifications and TLB flushes?
> >
> > Looking at the existing implementations of lazy MMU:
> >
> >  - In Xen PV implementation of lazy MMU, I see that TLB flush
> > hypercalls are delayed as well (xen_flush_tlb(),
> > xen_flush_tlb_one_user() and xen_flush_tlb_multi() all use
> > xen_mc_issue(XEN_LAZY_MMU) which delays issuing if lazymmu is active).
> >  - The sparc version also seems to delay TLB flushes, and sparc's
> > arch_leave_lazy_mmu_mode() seems to do TLB flushes via
> > flush_tlb_pending() if necessary.
> >  - powerpc's arch_leave_lazy_mmu_mode() also seems to do TLB flushes.
> >
> > Am I missing something?
>
> I doubt it. I suspect this was just my misunderstanding then. I hadn't
> appreciated that lazy mmu is also guarranteed to maintain flush ordering; it's
> chronically under-documented. Sorry for the noise here. On that basis, I 
> expect
> the first 2 patches can definitely be dropped.

Yeah looking at this code I agree that it could use significantly more
verbose comments on the API contract.