lkml.org 
[lkml]   [2017]   [Feb]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [PATCH] mprotect: drop overprotective lock_pte_protection()
    On Tue, Feb 07, 2017 at 05:33:47PM +0300, Kirill A. Shutemov wrote:
    > lock_pte_protection() uses pmd_lock() to make sure that we have stable
    > PTE page table before walking pte range.
    >
    > That's not necessary. We only need to make sure that PTE page table is
    > established. It cannot vanish under us as long as we hold mmap_sem at
    > least for read.
    >
    > And we already have helper for that -- pmd_trans_unstable().
    >
    > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
    > ---
    > mm/mprotect.c | 43 ++++++++++++-------------------------------
    > 1 file changed, 12 insertions(+), 31 deletions(-)
    >
    > diff --git a/mm/mprotect.c b/mm/mprotect.c
    > index f9c07f54dd62..e919e4613eab 100644
    > --- a/mm/mprotect.c
    > +++ b/mm/mprotect.c
    > @@ -33,34 +33,6 @@
    >
    > #include "internal.h"
    >
    > -/*
    > - * For a prot_numa update we only hold mmap_sem for read so there is a
    > - * potential race with faulting where a pmd was temporarily none. This
    > - * function checks for a transhuge pmd under the appropriate lock. It
    > - * returns a pte if it was successfully locked or NULL if it raced with
    > - * a transhuge insertion.
    > - */
    > -static pte_t *lock_pte_protection(struct vm_area_struct *vma, pmd_t *pmd,
    > - unsigned long addr, int prot_numa, spinlock_t **ptl)
    > -{
    > - pte_t *pte;
    > - spinlock_t *pmdl;
    > -
    > - /* !prot_numa is protected by mmap_sem held for write */
    > - if (!prot_numa)
    > - return pte_offset_map_lock(vma->vm_mm, pmd, addr, ptl);
    > -
    > - pmdl = pmd_lock(vma->vm_mm, pmd);
    > - if (unlikely(pmd_trans_huge(*pmd) || pmd_none(*pmd))) {
    > - spin_unlock(pmdl);
    > - return NULL;
    > - }
    > -
    > - pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, ptl);
    > - spin_unlock(pmdl);
    > - return pte;
    > -}
    > -
    > static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
    > unsigned long addr, unsigned long end, pgprot_t newprot,
    > int dirty_accountable, int prot_numa)
    > @@ -71,7 +43,7 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
    > unsigned long pages = 0;
    > int target_node = NUMA_NO_NODE;
    >
    > - pte = lock_pte_protection(vma, pmd, addr, prot_numa, &ptl);
    > + pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
    > if (!pte)

    I cleaned it up too but I moved the pmd_trans_unstable in the caller
    above instead of the callee, otherwise it's the same.

    >
    > @@ -177,8 +149,6 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
    > if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) {
    > if (next - addr != HPAGE_PMD_SIZE) {
    > __split_huge_pmd(vma, pmd, addr, false, NULL);
    > - if (pmd_trans_unstable(pmd))
    > - continue;

    Agree it can be removed too, but I only removed lock_pte_protection in
    my version.

    If you prefer this version to be merged so we don't have to cleanup
    the above superfluous check in a incremental patch that's fine of
    course, otherwise at runtime they're equivalent as far as I can
    tell. The version in -mm is here.

    https://git.kernel.org/cgit/linux/kernel/git/mhocko/mm.git/commit/?h=auto-latest&id=d84ff4e4985f397ca4ecfe7ec029c45c6f2b9906

    Thanks,
    Andrea

    \
     
     \ /
      Last update: 2017-02-07 22:29    [W:4.017 / U:0.116 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site