Re: [PATCH v3 2/2] mm: fix double page fault on arm64 if PTE_AF is cleared

[Kirill A. Shutemov]

On Mon, Sep 16, 2019 at 09:35:21AM +0000, Justin He (Arm Technology China) wrote:
> 
> Hi Kirill
> > -----Original Message-----
> > From: Kirill A. Shutemov <kirill@xxxxxxxxxxxxx>
> > Sent: 2019å9æ16æ 17:16
> > To: Justin He (Arm Technology China) <Justin.He@xxxxxxx>
> > Cc: Catalin Marinas <Catalin.Marinas@xxxxxxx>; Will Deacon
> > <will@xxxxxxxxxx>; Mark Rutland <Mark.Rutland@xxxxxxx>; James Morse
> > <James.Morse@xxxxxxx>; Marc Zyngier <maz@xxxxxxxxxx>; Matthew
> > Wilcox <willy@xxxxxxxxxxxxx>; Kirill A. Shutemov
> > <kirill.shutemov@xxxxxxxxxxxxxxx>; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx;
> > linux-kernel@xxxxxxxxxxxxxxx; linux-mm@xxxxxxxxx; Punit Agrawal
> > <punitagrawal@xxxxxxxxx>; Anshuman Khandual
> > <Anshuman.Khandual@xxxxxxx>; Jun Yao <yaojun8558363@xxxxxxxxx>;
> > Alex Van Brunt <avanbrunt@xxxxxxxxxx>; Robin Murphy
> > <Robin.Murphy@xxxxxxx>; Thomas Gleixner <tglx@xxxxxxxxxxxxx>;
> > Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>; JÃrÃme Glisse
> > <jglisse@xxxxxxxxxx>; Ralph Campbell <rcampbell@xxxxxxxxxx>;
> > hejianet@xxxxxxxxx
> > Subject: Re: [PATCH v3 2/2] mm: fix double page fault on arm64 if PTE_AF
> > is cleared
> >
> > On Sat, Sep 14, 2019 at 12:32:39AM +0800, Jia He wrote:
> > > When we tested pmdk unit test [1] vmmalloc_fork TEST1 in arm64 guest,
> > there
> > > will be a double page fault in __copy_from_user_inatomic of
> > cow_user_page.
> > >
> > > Below call trace is from arm64 do_page_fault for debugging purpose
> > > [  110.016195] Call trace:
> > > [  110.016826]  do_page_fault+0x5a4/0x690
> > > [  110.017812]  do_mem_abort+0x50/0xb0
> > > [  110.018726]  el1_da+0x20/0xc4
> > > [  110.019492]  __arch_copy_from_user+0x180/0x280
> > > [  110.020646]  do_wp_page+0xb0/0x860
> > > [  110.021517]  __handle_mm_fault+0x994/0x1338
> > > [  110.022606]  handle_mm_fault+0xe8/0x180
> > > [  110.023584]  do_page_fault+0x240/0x690
> > > [  110.024535]  do_mem_abort+0x50/0xb0
> > > [  110.025423]  el0_da+0x20/0x24
> > >
> > > The pte info before __copy_from_user_inatomic is (PTE_AF is cleared):
> > > [ffff9b007000] pgd=000000023d4f8003, pud=000000023da9b003,
> > pmd=000000023d4b3003, pte=360000298607bd3
> > >
> > > As told by Catalin: "On arm64 without hardware Access Flag, copying
> > from
> > > user will fail because the pte is old and cannot be marked young. So we
> > > always end up with zeroed page after fork() + CoW for pfn mappings. we
> > > don't always have a hardware-managed access flag on arm64."
> > >
> > > This patch fix it by calling pte_mkyoung. Also, the parameter is
> > > changed because vmf should be passed to cow_user_page()
> > >
> > > [1]
> > https://github.com/pmem/pmdk/tree/master/src/test/vmmalloc_fork
> > >
> > > Reported-by: Yibo Cai <Yibo.Cai@xxxxxxx>
> > > Signed-off-by: Jia He <justin.he@xxxxxxx>
> > > ---
> > >  mm/memory.c | 30 +++++++++++++++++++++++++-----
> > >  1 file changed, 25 insertions(+), 5 deletions(-)
> > >
> > > diff --git a/mm/memory.c b/mm/memory.c
> > > index e2bb51b6242e..a64af6495f71 100644
> > > --- a/mm/memory.c
> > > +++ b/mm/memory.c
> > > @@ -118,6 +118,13 @@ int randomize_va_space __read_mostly =
> > >                                     2;
> > >  #endif
> > >
> > > +#ifndef arch_faults_on_old_pte
> > > +static inline bool arch_faults_on_old_pte(void)
> > > +{
> > > +   return false;
> > > +}
> > > +#endif
> > > +
> > >  static int __init disable_randmaps(char *s)
> > >  {
> > >     randomize_va_space = 0;
> > > @@ -2140,7 +2147,8 @@ static inline int pte_unmap_same(struct
> > mm_struct *mm, pmd_t *pmd,
> > >     return same;
> > >  }
> > >
> > > -static inline void cow_user_page(struct page *dst, struct page *src,
> > unsigned long va, struct vm_area_struct *vma)
> > > +static inline void cow_user_page(struct page *dst, struct page *src,
> > > +                           struct vm_fault *vmf)
> > >  {
> > >     debug_dma_assert_idle(src);
> > >
> > > @@ -2152,20 +2160,32 @@ static inline void cow_user_page(struct page
> > *dst, struct page *src, unsigned lo
> > >      */
> > >     if (unlikely(!src)) {
> > >             void *kaddr = kmap_atomic(dst);
> > > -           void __user *uaddr = (void __user *)(va & PAGE_MASK);
> > > +           void __user *uaddr = (void __user *)(vmf->address &
> > PAGE_MASK);
> > > +           pte_t entry;
> > >
> > >             /*
> > >              * This really shouldn't fail, because the page is there
> > >              * in the page tables. But it might just be unreadable,
> > >              * in which case we just give up and fill the result with
> > > -            * zeroes.
> > > +            * zeroes. If PTE_AF is cleared on arm64, it might
> > > +            * cause double page fault. So makes pte young here
> > >              */
> > > +           if (arch_faults_on_old_pte() && !pte_young(vmf->orig_pte))
> > {
> > > +                   spin_lock(vmf->ptl);
> > > +                   entry = pte_mkyoung(vmf->orig_pte);
> >
> > Should't you re-validate that orig_pte after re-taking ptl? It can be
> > stale by now.
> Thanks, do you mean flush_cache_page(vma, vmf->address, pte_pfn(vmf->orig_pte))
> before pte_mkyoung?

No. You need to check pte_same(*vmf->pte, vmf->orig_pte) before modifying
anything and bail out if *vmf->pte has changed under you.

-- 
 Kirill A. Shutemov