[stable 5.10 PATCH] mm: khugepaged: skip huge page collapse for special files

From: Yang Shi
Date: Wed Nov 03 2021 - 16:23:05 EST


commit a4aeaa06d45e90f9b279f0b09de84bd00006e733 upstream.

The read-only THP for filesystems will collapse THP for files opened
readonly and mapped with VM_EXEC. The intended usecase is to avoid TLB
misses for large text segments. But it doesn't restrict the file types
so a THP could be collapsed for a non-regular file, for example, block
device, if it is opened readonly and mapped with EXEC permission. This
may cause bugs, like [1] and [2].

This is definitely not the intended usecase, so just collapse THP for
regular files in order to close the attack surface.

[shy828301@xxxxxxxxx: fix vm_file check [3]]

Link: https://lore.kernel.org/lkml/CACkBjsYwLYLRmX8GpsDpMthagWOjWWrNxqY6ZLNQVr6yx+f5vA@xxxxxxxxxxxxxx/ [1]
Link: https://lore.kernel.org/linux-mm/000000000000c6a82505ce284e4c@xxxxxxxxxx/ [2]
Link: https://lkml.kernel.org/r/CAHbLzkqTW9U3VvTu1Ki5v_cLRC9gHW+znBukg_ycergE0JWj-A@xxxxxxxxxxxxxx [3]
Link: https://lkml.kernel.org/r/20211027195221.3825-1-shy828301@xxxxxxxxx
Fixes: 99cb0dbd47a1 ("mm,thp: add read-only THP support for (non-shmem) FS")
Signed-off-by: Hugh Dickins <hughd@xxxxxxxxxx>
Signed-off-by: Yang Shi <shy828301@xxxxxxxxx>
Reported-by: Hao Sun <sunhao.th@xxxxxxxxx>
Reported-by: syzbot+aae069be1de40fb11825@xxxxxxxxxxxxxxxxxxxxxxxxx
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Song Liu <songliubraving@xxxxxx>
Cc: Andrea Righi <andrea.righi@xxxxxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
---
mm/khugepaged.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index ee8812578563..5c36848022de 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -443,21 +443,24 @@ static bool hugepage_vma_check(struct vm_area_struct *vma,
if (!transhuge_vma_enabled(vma, vm_flags))
return false;

+ if (vma->vm_file && !IS_ALIGNED((vma->vm_start >> PAGE_SHIFT) -
+ vma->vm_pgoff, HPAGE_PMD_NR))
+ return false;
+
/* Enabled via shmem mount options or sysfs settings. */
- if (shmem_file(vma->vm_file) && shmem_huge_enabled(vma)) {
- return IS_ALIGNED((vma->vm_start >> PAGE_SHIFT) - vma->vm_pgoff,
- HPAGE_PMD_NR);
- }
+ if (shmem_file(vma->vm_file))
+ return shmem_huge_enabled(vma);

/* THP settings require madvise. */
if (!(vm_flags & VM_HUGEPAGE) && !khugepaged_always())
return false;

- /* Read-only file mappings need to be aligned for THP to work. */
+ /* Only regular file is valid */
if (IS_ENABLED(CONFIG_READ_ONLY_THP_FOR_FS) && vma->vm_file &&
(vm_flags & VM_DENYWRITE)) {
- return IS_ALIGNED((vma->vm_start >> PAGE_SHIFT) - vma->vm_pgoff,
- HPAGE_PMD_NR);
+ struct inode *inode = vma->vm_file->f_inode;
+
+ return S_ISREG(inode->i_mode);
}

if (!vma->anon_vma || vma->vm_ops)
--
2.26.2