> On Jul 30, 2019, at 7:59 AM, Kirill A. Shutemov <kirill@xxxxxxxxxxxxx> wrote:
>
> On Sun, Jul 28, 2019 at 10:43:34PM -0700, Song Liu wrote:
>> khugepaged needs exclusive mmap_sem to access page table. When it fails
>> to lock mmap_sem, the page will fault in as pte-mapped THP. As the page
>> is already a THP, khugepaged will not handle this pmd again.
>>
>> This patch enables the khugepaged to retry collapse the page table.
>>
>> struct mm_slot (in khugepaged.c) is extended with an array, containing
>> addresses of pte-mapped THPs. We use array here for simplicity. We can
>> easily replace it with more advanced data structures when needed. This
>> array is protected by khugepaged_mm_lock.
>>
>> In khugepaged_scan_mm_slot(), if the mm contains pte-mapped THP, we try
>> to collapse the page table.
>>
>> Since collapse may happen at an later time, some pages may already fault
>> in. collapse_pte_mapped_thp() is added to properly handle these pages.
>> collapse_pte_mapped_thp() also double checks whether all ptes in this pmd
>> are mapping to the same THP. This is necessary because some subpage of
>> the THP may be replaced, for example by uprobe. In such cases, it is not
>> possible to collapse the pmd.
>>
>> Signed-off-by: Song Liu <songliubraving@xxxxxx>
>> ---
>> include/linux/khugepaged.h | 15 ++++
>> mm/khugepaged.c | 136 +++++++++++++++++++++++++++++++++++++
>> 2 files changed, 151 insertions(+)
>>
>> diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
>> index 082d1d2a5216..2d700830fe0e 100644
>> --- a/include/linux/khugepaged.h
>> +++ b/include/linux/khugepaged.h
>> @@ -15,6 +15,16 @@ extern int __khugepaged_enter(struct mm_struct *mm);
>> extern void __khugepaged_exit(struct mm_struct *mm);
>> extern int khugepaged_enter_vma_merge(struct vm_area_struct *vma,
>> unsigned long vm_flags);
>> +#ifdef CONFIG_SHMEM
>> +extern int khugepaged_add_pte_mapped_thp(struct mm_struct *mm,
>> + unsigned long addr);
>> +#else
>> +static inline int khugepaged_add_pte_mapped_thp(struct mm_struct *mm,
>> + unsigned long addr)
>> +{
>> + return 0;
>> +}
>> +#endif
>>
>> #define khugepaged_enabled() \
>> (transparent_hugepage_flags & \
>> @@ -73,6 +83,11 @@ static inline int khugepaged_enter_vma_merge(struct vm_area_struct *vma,
>> {
>> return 0;
>> }
>> +static inline int khugepaged_add_pte_mapped_thp(struct mm_struct *mm,
>> + unsigned long addr)
>> +{
>> + return 0;
>> +}
>> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>>
>> #endif /* _LINUX_KHUGEPAGED_H */
>> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
>> index eaaa21b23215..247c25aeb096 100644
>> --- a/mm/khugepaged.c
>> +++ b/mm/khugepaged.c
>> @@ -76,6 +76,7 @@ static __read_mostly DEFINE_HASHTABLE(mm_slots_hash, MM_SLOTS_HASH_BITS);
>>
>> static struct kmem_cache *mm_slot_cache __read_mostly;
>>
>> +#define MAX_PTE_MAPPED_THP 8
>
> Is MAX_PTE_MAPPED_THP value random or do you have any justification for
> it?
In our use cases, we only have small number (< 10) of huge pages for the
text section, so 8 should be enough to cover the worse case.
If this is not sufficient, we can make it a list.
>
> Please add empty line after it.
>
>> /**
>> * struct mm_slot - hash lookup from mm to mm_slot
>> * @hash: hash collision list
>> @@ -86,6 +87,10 @@ struct mm_slot {
>> struct hlist_node hash;
>> struct list_head mm_node;
>> struct mm_struct *mm;
>> +
>> + /* pte-mapped THP in this mm */
>> + int nr_pte_mapped_thp;
>> + unsigned long pte_mapped_thp[MAX_PTE_MAPPED_THP];
>> };
>>
>> /**
>> @@ -1281,11 +1286,141 @@ static void retract_page_tables(struct address_space *mapping, pgoff_t pgoff)
>> up_write(&vma->vm_mm->mmap_sem);
>> mm_dec_nr_ptes(vma->vm_mm);
>> pte_free(vma->vm_mm, pmd_pgtable(_pmd));
>> + } else if (down_read_trylock(&vma->vm_mm->mmap_sem)) {
>> + /* need down_read for khugepaged_test_exit() */
>> + khugepaged_add_pte_mapped_thp(vma->vm_mm, addr);
>> + up_read(&vma->vm_mm->mmap_sem);
>> }
>> }
>> i_mmap_unlock_write(mapping);
>> }
>>
>> +/*
>> + * Notify khugepaged that given addr of the mm is pte-mapped THP. Then
>> + * khugepaged should try to collapse the page table.
>> + */
>> +int khugepaged_add_pte_mapped_thp(struct mm_struct *mm, unsigned long addr)
>
> What is contract about addr alignment? Do we expect it PAGE_SIZE aligned
> or PMD_SIZE aligned? Do we want to enforce it?
It is PMD_SIZE aligned. Let me add VM_BUG_ON() for it.
>
>> +{
>> + struct mm_slot *mm_slot;
>> + int ret = 0;
>> +
>> + /* hold mmap_sem for khugepaged_test_exit() */
>> + VM_BUG_ON_MM(!rwsem_is_locked(&mm->mmap_sem), mm);
>> +
>> + if (unlikely(khugepaged_test_exit(mm)))
>> + return 0;
>> +
>> + if (!test_bit(MMF_VM_HUGEPAGE, &mm->flags) &&
>> + !test_bit(MMF_DISABLE_THP, &mm->flags)) {
>> + ret = __khugepaged_enter(mm);
>> + if (ret)
>> + return ret;
>> + }
>
> Any reason not to call khugepaged_enter() here?
No specific reasons... Let me try it.
>
>> +
>> + spin_lock(&khugepaged_mm_lock);
>> + mm_slot = get_mm_slot(mm);
>> + if (likely(mm_slot && mm_slot->nr_pte_mapped_thp < MAX_PTE_MAPPED_THP))
>> + mm_slot->pte_mapped_thp[mm_slot->nr_pte_mapped_thp++] = addr;
>
> It's probably good enough for start, but I'm not sure how useful it will
> be for real application, considering the limitation.
For limitation, do you mean MAX_PTE_MAPPED_THP? I think this is good for
our use cases. We sure can improve that without too much work.
Thanks,
Song
>
>> +
>
> Useless empty line?
>
>> + spin_unlock(&khugepaged_mm_lock);
>> + return 0;
>> +}
>> +
>> +/**
>> + * Try to collapse a pte-mapped THP for mm at address haddr.
>> + *
>> + * This function checks whether all the PTEs in the PMD are pointing to the
>> + * right THP. If so, retract the page table so the THP can refault in with
>> + * as pmd-mapped.
>> + */
>> +static void collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long haddr)
>> +{
>> + struct vm_area_struct *vma = find_vma(mm, haddr);
>> + pmd_t *pmd = mm_find_pmd(mm, haddr);
>> + struct page *hpage = NULL;
>> + unsigned long addr;
>> + spinlock_t *ptl;
>> + int count = 0;
>> + pmd_t _pmd;
>> + int i;
>> +
>> + if (!vma || !pmd || pmd_trans_huge(*pmd))
>> + return;
>> +
>> + /* step 1: check all mapped PTEs are to the right huge page */
>> + for (i = 0, addr = haddr; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE) {
>> + pte_t *pte = pte_offset_map(pmd, addr);
>> + struct page *page;
>> +
>> + if (pte_none(*pte))
>> + continue;
>> +
>> + page = vm_normal_page(vma, addr, *pte);
>> +
>> + if (!PageCompound(page))
>> + return;
>
> I think khugepaged_scan_shmem() and collapse_shmem() should changed to not
> stop on PageTransCompound() to make this useful for more cases.
With locking sorted out, we could call collapse_pte_mapped_thp() for
PageTransCompound() cases.
>
> Ideally, it collapse_shmem() and this routine should be the same thing.
> Or do you thing it's not doable for some reason?
This routine is part of retract_page_tables(). collapse_shmem() does more
work. We can still go into collapse_shmem() and bypass the first half of
it.
On the other hand, I would like to keep it separate for now to minimize
conflicts with my other set, which is waiting for fixed version of
5fd4ca2d84b249f0858ce28cf637cf25b61a398f.
How about we start with current design (with small things fixed) and
improve it once both sets get in?
Thanks,
Song
