Hey Zach,
Thanks for taking time to review!
On Tue, Jan 30, 2024 at 3:04 AM Zach O'Keefe <zokeefe@xxxxxxxxxx> wrote:
[...]
> IIUC, there really isn't any correctness race. Claim is just that we
Yes, there is indeed no correctness race.
> can avoid a number of per-vma checks. AFAICT, any task w/
> MMF_DISABLE_THP set will always have each and every vma checked
> (albeit, with a very inexpensive ->vm_mm->flags check)
[...]
IMO, for any task with MMF_DISABLE_THP set, the check
for each VMA can be skipped to avoid redundant operations,
(with a very inexpensive ->mm->flags check)
especially in scenarios with a large address space.
BR,
Lance
On Tue, Jan 30, 2024 at 3:04 AM Zach O'Keefe <zokeefe@xxxxxxxxxx> wrote:
>
> On Mon, Jan 29, 2024 at 10:53 AM Yang Shi <shy828301@xxxxxxxxx> wrote:
> >
> > On Sun, Jan 28, 2024 at 9:46 PM Lance Yang <ioworker0@xxxxxxxxx> wrote:
> > >
> > > khugepaged scans the entire address space in the
> > > background for each given mm, looking for
> > > opportunities to merge sequences of basic pages
> > > into huge pages. However, when an mm is inserted
> > > to the mm_slots list, and the MMF_DISABLE_THP flag
> > > is set later, this scanning process becomes
> > > unnecessary for that mm and can be skipped to avoid
> > > redundant operations, especially in scenarios with
> > > a large address space.
> > >
> > > This commit introduces a check before each scanning
> > > process to test the MMF_DISABLE_THP flag for the
> > > given mm; if the flag is set, the scanning process
> > > is bypassed, thereby improving the efficiency of
> > > khugepaged.
> > >
> > > Signed-off-by: Lance Yang <ioworker0@xxxxxxxxx>
> > > ---
> > > mm/khugepaged.c | 18 ++++++++++++------
> > > 1 file changed, 12 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> > > index 2b219acb528e..d6a700834edc 100644
> > > --- a/mm/khugepaged.c
> > > +++ b/mm/khugepaged.c
> > > @@ -410,6 +410,12 @@ static inline int hpage_collapse_test_exit(struct mm_struct *mm)
> > > return atomic_read(&mm->mm_users) == 0;
> > > }
> > >
> > > +static inline int hpage_collapse_test_exit_or_disable(struct mm_struct *mm)
> > > +{
> > > + return hpage_collapse_test_exit(mm) ||
> > > + test_bit(MMF_DISABLE_THP, &mm->flags);
> > > +}
> > > +
> > > void __khugepaged_enter(struct mm_struct *mm)
> > > {
> > > struct khugepaged_mm_slot *mm_slot;
> > > @@ -1422,7 +1428,7 @@ static void collect_mm_slot(struct khugepaged_mm_slot *mm_slot)
> > >
> > > lockdep_assert_held(&khugepaged_mm_lock);
> > >
> > > - if (hpage_collapse_test_exit(mm)) {
> > > + if (hpage_collapse_test_exit_or_disable(mm)) {
> > > /* free mm_slot */
> > > hash_del(&slot->hash);
> > > list_del(&slot->mm_node);
> > > @@ -2360,7 +2366,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
> > > goto breakouterloop_mmap_lock;
> > >
> > > progress++;
> > > - if (unlikely(hpage_collapse_test_exit(mm)))
> > > + if (unlikely(hpage_collapse_test_exit_or_disable(mm)))
> > > goto breakouterloop;
> > >
> > > vma_iter_init(&vmi, mm, khugepaged_scan.address);
> > > @@ -2368,7 +2374,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
> > > unsigned long hstart, hend;
> > >
> > > cond_resched();
> > > - if (unlikely(hpage_collapse_test_exit(mm))) {
> > > + if (unlikely(hpage_collapse_test_exit_or_disable(mm))) {
> >
> > The later thp_vma_allowable_order() does check whether MMF_DISABLE_THP
> > is set or not. And the hugepage_vma_revalidate() after re-acquiring
> > mmap_lock does the same check too. The checking in khugepaged should
> > be already serialized with prctl, which takes mmap_lock in write.
>
> IIUC, there really isn't any correctness race. Claim is just that we
> can avoid a number of per-vma checks. AFAICT, any task w/
> MMF_DISABLE_THP set will always have each and every vma checked
> (albeit, with a very inexpensive ->vm_mm->flags check)
>
> Thanks,
> Zach
>
> > > progress++;
> > > break;
> > > }
> > > @@ -2390,7 +2396,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
> > > bool mmap_locked = true;
> > >
> > > cond_resched();
> > > - if (unlikely(hpage_collapse_test_exit(mm)))
> > > + if (unlikely(hpage_collapse_test_exit_or_disable(mm)))
> > > goto breakouterloop;
> > >
> > > VM_BUG_ON(khugepaged_scan.address < hstart ||
> > > @@ -2408,7 +2414,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
> > > fput(file);
> > > if (*result == SCAN_PTE_MAPPED_HUGEPAGE) {
> > > mmap_read_lock(mm);
> > > - if (hpage_collapse_test_exit(mm))
> > > + if (hpage_collapse_test_exit_or_disable(mm))
> > > goto breakouterloop;
> > > *result = collapse_pte_mapped_thp(mm,
> > > khugepaged_scan.address, false);
> > > @@ -2450,7 +2456,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
> > > * Release the current mm_slot if this mm is about to die, or
> > > * if we scanned all vmas of this mm.
> > > */
> > > - if (hpage_collapse_test_exit(mm) || !vma) {
> > > + if (hpage_collapse_test_exit_or_disable(mm) || !vma) {
> > > /*
> > > * Make sure that if mm_users is reaching zero while
> > > * khugepaged runs here, khugepaged_exit will find
> > > --
> > > 2.33.1
> > >
