From: Hugh Dickins hughd@google.com
[ Upstream commit a9a7504d9beaf395481faa91e70e2fd08f7a3dde ]
Running certain tests with a DEBUG_VM kernel would crash within hours, on the total_mapcount BUG() in split_huge_page_to_list(), while trying to free up some memory by punching a hole in a shmem huge page: split's try_to_unmap() was unable to find all the mappings of the page (which, on a !DEBUG_VM kernel, would then keep the huge page pinned in memory).
Crash dumps showed two tail pages of a shmem huge page remained mapped by pte: ptes in a non-huge-aligned vma of a gVisor process, at the end of a long unmapped range; and no page table had yet been allocated for the head of the huge page to be mapped into.
Although designed to handle these odd misaligned huge-page-mapped-by-pte cases, page_vma_mapped_walk() falls short by returning false prematurely when !pmd_present or !pud_present or !p4d_present or !pgd_present: there are cases when a huge page may span the boundary, with ptes present in the next.
Restructure page_vma_mapped_walk() as a loop to continue in these cases, while keeping its layout much as before. Add a step_forward() helper to advance pvmw->address across those boundaries: originally I tried to use mm's standard p?d_addr_end() macros, but hit the same crash 512 times less often: because of the way redundant levels are folded together, but folded differently in different configurations, it was just too difficult to use them correctly; and step_forward() is simpler anyway.
Link: https://lkml.kernel.org/r/fedb8632-1798-de42-f39e-873551d5bc81@google.com Fixes: ace71a19cec5 ("mm: introduce page_vma_mapped_walk()") Signed-off-by: Hugh Dickins hughd@google.com Acked-by: Kirill A. Shutemov kirill.shutemov@linux.intel.com Cc: Alistair Popple apopple@nvidia.com Cc: Matthew Wilcox willy@infradead.org Cc: Peter Xu peterx@redhat.com Cc: Ralph Campbell rcampbell@nvidia.com Cc: Wang Yugui wangyugui@e16-tech.com Cc: Will Deacon will@kernel.org Cc: Yang Shi shy828301@gmail.com Cc: Zi Yan ziy@nvidia.com Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Linus Torvalds torvalds@linux-foundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- mm/page_vma_mapped.c | 34 +++++++++++++++++++++++++--------- 1 file changed, 25 insertions(+), 9 deletions(-)
diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c index 7239c38e99e2..b7a8009c1549 100644 --- a/mm/page_vma_mapped.c +++ b/mm/page_vma_mapped.c @@ -111,6 +111,13 @@ static bool check_pte(struct page_vma_mapped_walk *pvmw) return pfn_in_hpage(pvmw->page, pfn); }
+static void step_forward(struct page_vma_mapped_walk *pvmw, unsigned long size) +{ + pvmw->address = (pvmw->address + size) & ~(size - 1); + if (!pvmw->address) + pvmw->address = ULONG_MAX; +} + /** * page_vma_mapped_walk - check if @pvmw->page is mapped in @pvmw->vma at * @pvmw->address @@ -179,16 +186,22 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw) if (pvmw->pte) goto next_pte; restart: - { + do { pgd = pgd_offset(mm, pvmw->address); - if (!pgd_present(*pgd)) - return false; + if (!pgd_present(*pgd)) { + step_forward(pvmw, PGDIR_SIZE); + continue; + } p4d = p4d_offset(pgd, pvmw->address); - if (!p4d_present(*p4d)) - return false; + if (!p4d_present(*p4d)) { + step_forward(pvmw, P4D_SIZE); + continue; + } pud = pud_offset(p4d, pvmw->address); - if (!pud_present(*pud)) - return false; + if (!pud_present(*pud)) { + step_forward(pvmw, PUD_SIZE); + continue; + }
pvmw->pmd = pmd_offset(pud, pvmw->address); /* @@ -235,7 +248,8 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
spin_unlock(ptl); } - return false; + step_forward(pvmw, PMD_SIZE); + continue; } if (!map_pte(pvmw)) goto next_pte; @@ -265,7 +279,9 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw) spin_lock(pvmw->ptl); } goto this_pte; - } + } while (pvmw->address < end); + + return false; }
/**