On Wed, Sep 2, 2020 at 9:58 AM Ralph Campbell <rcampb...@nvidia.com> wrote:
>
> A migrating transparent huge page has to already be unmapped. Otherwise,
> the page could be modified while it is being copied to a new page and
> data could be lost. The function __split_huge_pmd() checks for a PMD
> migration entry before calling __split_huge_pmd_locked() leading one to
> think that __split_huge_pmd_locked() can handle splitting a migrating PMD.
> However, the code always increments the page->_mapcount and adjusts the
> memory control group accounting assuming the page is mapped.
> Also, if the PMD entry is a migration PMD entry, the call to
> is_huge_zero_pmd(*pmd) is incorrect because it calls pmd_pfn(pmd) instead
> of migration_entry_to_pfn(pmd_to_swp_entry(pmd)).
> Fix these problems by checking for a PMD migration entry.
Thanks for catching this. The fix looks good to me. Reviewed-by: Yang
Shi <shy828...@gmail.com>
I think this fix can go separately with the series.
>
> Signed-off-by: Ralph Campbell <rcampb...@nvidia.com>
> ---
> mm/huge_memory.c | 42 +++++++++++++++++++++++-------------------
> 1 file changed, 23 insertions(+), 19 deletions(-)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 2a468a4acb0a..606d712d9505 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -2023,7 +2023,7 @@ static void __split_huge_pmd_locked(struct
> vm_area_struct *vma, pmd_t *pmd,
> put_page(page);
> add_mm_counter(mm, mm_counter_file(page), -HPAGE_PMD_NR);
> return;
> - } else if (is_huge_zero_pmd(*pmd)) {
> + } else if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
> /*
> * FIXME: Do we want to invalidate secondary mmu by calling
> * mmu_notifier_invalidate_range() see comments below inside
> @@ -2117,30 +2117,34 @@ static void __split_huge_pmd_locked(struct
> vm_area_struct *vma, pmd_t *pmd,
> pte = pte_offset_map(&_pmd, addr);
> BUG_ON(!pte_none(*pte));
> set_pte_at(mm, addr, pte, entry);
> - atomic_inc(&page[i]._mapcount);
> - pte_unmap(pte);
> - }
> -
> - /*
> - * Set PG_double_map before dropping compound_mapcount to avoid
> - * false-negative page_mapped().
> - */
> - if (compound_mapcount(page) > 1 && !TestSetPageDoubleMap(page)) {
> - for (i = 0; i < HPAGE_PMD_NR; i++)
> + if (!pmd_migration)
> atomic_inc(&page[i]._mapcount);
> + pte_unmap(pte);
> }
>
> - lock_page_memcg(page);
> - if (atomic_add_negative(-1, compound_mapcount_ptr(page))) {
> - /* Last compound_mapcount is gone. */
> - __dec_lruvec_page_state(page, NR_ANON_THPS);
> - if (TestClearPageDoubleMap(page)) {
> - /* No need in mapcount reference anymore */
> + if (!pmd_migration) {
> + /*
> + * Set PG_double_map before dropping compound_mapcount to
> avoid
> + * false-negative page_mapped().
> + */
> + if (compound_mapcount(page) > 1 &&
> + !TestSetPageDoubleMap(page)) {
> for (i = 0; i < HPAGE_PMD_NR; i++)
> - atomic_dec(&page[i]._mapcount);
> + atomic_inc(&page[i]._mapcount);
> + }
> +
> + lock_page_memcg(page);
> + if (atomic_add_negative(-1, compound_mapcount_ptr(page))) {
> + /* Last compound_mapcount is gone. */
> + __dec_lruvec_page_state(page, NR_ANON_THPS);
> + if (TestClearPageDoubleMap(page)) {
> + /* No need in mapcount reference anymore */
> + for (i = 0; i < HPAGE_PMD_NR; i++)
> + atomic_dec(&page[i]._mapcount);
> + }
> }
> + unlock_page_memcg(page);
> }
> - unlock_page_memcg(page);
>
> smp_wmb(); /* make pte visible before pmd */
> pmd_populate(mm, pmd, pgtable);
> --
> 2.20.1
>
>
_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau
