mm/page_table_check: skip special zero mappings

page_table_check_set() and page_table_check_clear() account mappings based
on PageAnon().  Shared zero-page PTEs and huge zero PMDs are special
mappings, but page_table_check can still account them as file-backed
pages.

An unprivileged process can populate enough zero mappings to overflow
file_map_count and hit the existing BUG_ON().  The PTE path can do this
with the shared zero page, and the PMD path can do the same with huge zero
mappings.

Skip special zero mappings in the user page-table accounting paths.  Keep
the PTE-side pte_special() check, and identify huge zero PMDs from the
mapped folio instead of pmd_special().  That covers architectures where
pmd_special() is a no-op without adding huge_zero_pfn checks to the
generic counter helpers.

Link: https://lore.kernel.org/cover.1784717203.git.zhilinz@nebusec.ai
Link: https://lore.kernel.org/e94478e4fb7912fb7e8ebebed5ce85d00dc9a69d.1784717203.git.zhilinz@nebusec.ai
Fixes: df4e817b7108 ("mm: page table check")
Signed-off-by: Zhiling Zou <zhilinz@nebusec.ai>
Signed-off-by: Ren Wei <enjou1224z@gmail.com>
Reported-by: Vega <vega@nebusec.ai>
Cc: Pasha Tatashin <pasha.tatashin@soleen.com>
Assisted-by: Codex:gpt-5.4
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/page_table_check.c b/mm/page_table_check.c
index 53a8997ec043ea9df38efc844057329a9e1e4e8e..2403f5a1141058e0fdf3117f1e5cf09f849c63c9 100644 (file)
--- a/mm/page_table_check.c
+++ b/mm/page_table_check.c
@@ -151,18 +151,29 @@ void __page_table_check_pte_clear(struct mm_struct *mm, unsigned long addr,
        if (&init_mm == mm)
                return;
 
-       if (pte_user_accessible_page(mm, addr, pte))
+       if (pte_user_accessible_page(mm, addr, pte) && !pte_special(pte))
                page_table_check_clear(pte_pfn(pte), PAGE_SIZE >> PAGE_SHIFT);
 }
 EXPORT_SYMBOL(__page_table_check_pte_clear);
 
+static inline bool page_table_check_huge_zero_pmd(pmd_t pmd)
+{
+       unsigned long pfn = pmd_pfn(pmd);
+
+       if (!pfn_valid(pfn))
+               return false;
+
+       return is_huge_zero_folio(page_folio(pfn_to_page(pfn)));
+}
+
 void __page_table_check_pmd_clear(struct mm_struct *mm, unsigned long addr,
                                  pmd_t pmd)
 {
        if (&init_mm == mm)
                return;
 
-       if (pmd_user_accessible_page(mm, addr, pmd))
+       if (pmd_user_accessible_page(mm, addr, pmd) &&
+           !page_table_check_huge_zero_pmd(pmd))
                page_table_check_clear(pmd_pfn(pmd), PMD_SIZE >> PAGE_SHIFT);
 }
 EXPORT_SYMBOL(__page_table_check_pmd_clear);
@@ -208,7 +219,7 @@ void __page_table_check_ptes_set(struct mm_struct *mm, unsigned long addr,
 
        for (i = 0; i < nr; i++)
                __page_table_check_pte_clear(mm, addr + PAGE_SIZE * i, ptep_get(ptep + i));
-       if (pte_user_accessible_page(mm, addr, pte))
+       if (pte_user_accessible_page(mm, addr, pte) && !pte_special(pte))
                page_table_check_set(pte_pfn(pte), nr, pte_write(pte));
 }
 EXPORT_SYMBOL(__page_table_check_ptes_set);
@@ -238,7 +249,8 @@ void __page_table_check_pmds_set(struct mm_struct *mm, unsigned long addr,
 
        for (i = 0; i < nr; i++)
                __page_table_check_pmd_clear(mm, addr + PMD_SIZE * i, *(pmdp + i));
-       if (pmd_user_accessible_page(mm, addr, pmd))
+       if (pmd_user_accessible_page(mm, addr, pmd) &&
+           !page_table_check_huge_zero_pmd(pmd))
                page_table_check_set(pmd_pfn(pmd), stride * nr, pmd_write(pmd));
 }
 EXPORT_SYMBOL(__page_table_check_pmds_set);