[PATCH v1 4/9] mm/memory: Add tree limit to free_pgtables()
Suren Baghdasaryan
surenb at google.com
Tue Sep 9 14:05:47 PDT 2025
On Tue, Sep 9, 2025 at 12:10 PM Liam R. Howlett <Liam.Howlett at oracle.com> wrote:
>
> The ceiling and tree search limit need to be different arguments for the
> future change in the failed fork attempt.
>
> Add some documentation around free_pgtables() and the limits in an
> attempt to clarify the floor and ceiling use as well as the new
> tree_max.
>
> Test code also updated.
>
> No functional changes intended.
>
> Signed-off-by: Liam R. Howlett <Liam.Howlett at oracle.com>
> ---
> mm/internal.h | 4 +++-
> mm/memory.c | 28 +++++++++++++++++++++++++---
> mm/mmap.c | 2 +-
> mm/vma.c | 3 ++-
> tools/testing/vma/vma_internal.h | 3 ++-
> 5 files changed, 33 insertions(+), 7 deletions(-)
>
> diff --git a/mm/internal.h b/mm/internal.h
> index 63e3ec8d63be7..d295252407fee 100644
> --- a/mm/internal.h
> +++ b/mm/internal.h
> @@ -444,7 +444,9 @@ void folio_activate(struct folio *folio);
>
> void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> struct vm_area_struct *start_vma, unsigned long floor,
> - unsigned long ceiling, bool mm_wr_locked);
> + unsigned long ceiling, unsigned long tree_max,
> + bool mm_wr_locked);
> +
> void pmd_install(struct mm_struct *mm, pmd_t *pmd, pgtable_t *pte);
>
> struct zap_details;
> diff --git a/mm/memory.c b/mm/memory.c
> index 3e0404bd57a02..24716b3713f66 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -369,12 +369,34 @@ void free_pgd_range(struct mmu_gather *tlb,
> } while (pgd++, addr = next, addr != end);
> }
>
> +/*
> + * free_pgtables() - Free a range of page tables
> + * @tlb: The mmu gather
> + * @mas: The maple state
> + * @vma: The first vma
> + * @floor: The lowest page table address
> + * @ceiling: The highest page table address
> + * @tree_max: The highest tree search address
> + * @mm_wr_locked: boolean indicating if the mm is write locked
> + *
> + * Note: Floor and ceiling are provided to indicate the absolute range of the
> + * page tables that should be removed. This can differ from the vma mappings on
> + * some archs that may have mappings that need to be removed outside the vmas.
> + * Note that the prev->vm_end and next->vm_start are often used.
> + *
> + * The tree_max differs from the ceiling when a dup_mmap() failed and the tree
> + * has unrelated data to the mm_struct being torn down.
> + */
> void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> struct vm_area_struct *vma, unsigned long floor,
> - unsigned long ceiling, bool mm_wr_locked)
> + unsigned long ceiling, unsigned long tree_max,
> + bool mm_wr_locked)
> {
> struct unlink_vma_file_batch vb;
>
> + /* underflow can happen and is fine */
This comment is a bit confusing... I think the below check covers 2 cases:
1. if ceiling == 0 then tree_max can be anything;
2. if ceiling > 0 then tree_max <= ceiling;
Is that what you intended? If so, maybe amend the comments for the
free_pgtables() function explaining this more explicitly?
> + WARN_ON_ONCE(tree_max - 1 > ceiling - 1);
I would prefer WARN_ON_ONCE(ceiling && tree_max > ceiling); as more
descriptive but that might be just me.
> +
> tlb_free_vmas(tlb);
>
> do {
> @@ -385,7 +407,7 @@ void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> * Note: USER_PGTABLES_CEILING may be passed as ceiling and may
You replaced ceiling with tree_max, so the above comment needs to be
updated as well.
> * be 0. This will underflow and is okay.
> */
> - next = mas_find(mas, ceiling - 1);
> + next = mas_find(mas, tree_max - 1);
> if (unlikely(xa_is_zero(next)))
> next = NULL;
>
> @@ -405,7 +427,7 @@ void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> */
> while (next && next->vm_start <= vma->vm_end + PMD_SIZE) {
> vma = next;
> - next = mas_find(mas, ceiling - 1);
> + next = mas_find(mas, tree_max - 1);
> if (unlikely(xa_is_zero(next)))
> next = NULL;
> if (mm_wr_locked)
> diff --git a/mm/mmap.c b/mm/mmap.c
> index a290448a53bb2..0f4808f135fe6 100644
> --- a/mm/mmap.c
> +++ b/mm/mmap.c
> @@ -1311,7 +1311,7 @@ void exit_mmap(struct mm_struct *mm)
> mt_clear_in_rcu(&mm->mm_mt);
> vma_iter_set(&vmi, vma->vm_end);
> free_pgtables(&tlb, &vmi.mas, vma, FIRST_USER_ADDRESS,
> - USER_PGTABLES_CEILING, true);
> + USER_PGTABLES_CEILING, USER_PGTABLES_CEILING, true);
> tlb_finish_mmu(&tlb);
>
> /*
> diff --git a/mm/vma.c b/mm/vma.c
> index a648e0555c873..1bae142bbc0f1 100644
> --- a/mm/vma.c
> +++ b/mm/vma.c
> @@ -486,6 +486,7 @@ void unmap_region(struct ma_state *mas, struct vm_area_struct *vma,
> /* mm_wr_locked = */ true);
> mas_set(mas, vma->vm_end);
> free_pgtables(&tlb, mas, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS,
> + next ? next->vm_start : USER_PGTABLES_CEILING,
> next ? next->vm_start : USER_PGTABLES_CEILING,
> /* mm_wr_locked = */ true);
> tlb_finish_mmu(&tlb);
> @@ -1232,7 +1233,7 @@ static inline void vms_clear_ptes(struct vma_munmap_struct *vms,
> mas_set(mas_detach, 1);
> /* start and end may be different if there is no prev or next vma. */
> free_pgtables(&tlb, mas_detach, vms->vma, vms->unmap_start,
> - vms->unmap_end, mm_wr_locked);
> + vms->unmap_end, vms->unmap_end, mm_wr_locked);
> tlb_finish_mmu(&tlb);
> vms->clear_ptes = false;
> }
> diff --git a/tools/testing/vma/vma_internal.h b/tools/testing/vma/vma_internal.h
> index 07167446dcf42..823d379e1fac2 100644
> --- a/tools/testing/vma/vma_internal.h
> +++ b/tools/testing/vma/vma_internal.h
> @@ -900,7 +900,8 @@ static inline void unmap_vmas(struct mmu_gather *tlb, struct ma_state *mas,
>
> static inline void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> struct vm_area_struct *vma, unsigned long floor,
> - unsigned long ceiling, bool mm_wr_locked)
> + unsigned long ceiling, unsigned long tree_max,
> + bool mm_wr_locked)
> {
> (void)tlb;
> (void)mas;
> --
> 2.47.2
>
More information about the maple-tree
mailing list