Re: [PATCH v7 2/4] mm,page_owner: Implement the tracking of the stacks count

[Marco Elver]

On Fri, 9 Feb 2024 at 00:45, Oscar Salvador <osalvador@xxxxxxx> wrote:
>
> page_owner needs to increment a stack_record refcount when a new allocation
> occurs, and decrement it on a free operation.
> In order to do that, we need to have a way to get a stack_record from a
> handle.
> Implement stack_depot_get_stack() which just does that, and make it public
> so page_owner can use it.
>
> Also implement {inc,dec}_stack_record_count() which increments
> or decrements on respective allocation and free operations, via
> __reset_page_owner() (free operation) and __set_page_owner() (alloc
> operation).
>
> Signed-off-by: Oscar Salvador <osalvador@xxxxxxx>
> ---
>  include/linux/stackdepot.h |  8 ++++++++
>  lib/stackdepot.c           |  8 ++++++++
>  mm/page_owner.c            | 30 ++++++++++++++++++++++++++++++
>  3 files changed, 46 insertions(+)
>
> diff --git a/include/linux/stackdepot.h b/include/linux/stackdepot.h
> index d0dcf4aebfb4..ac62de4d4999 100644
> --- a/include/linux/stackdepot.h
> +++ b/include/linux/stackdepot.h
> @@ -175,6 +175,14 @@ depot_stack_handle_t stack_depot_save_flags(unsigned long *entries,
>  depot_stack_handle_t stack_depot_save(unsigned long *entries,
>                                       unsigned int nr_entries, gfp_t gfp_flags);
>
> +/**
> + * stack_depo_get_stack - Get a pointer to a stack struct
> + * @handle: Stack depot handle
> + *
> + * Return: Returns a pointer to a stack struct
> + */



> +struct stack_record *stack_depot_get_stack(depot_stack_handle_t handle);
> +
>  /**
>   * stack_depot_fetch - Fetch a stack trace from stack depot
>   *
> diff --git a/lib/stackdepot.c b/lib/stackdepot.c
> index 16c8a1bf0008..197c355601f9 100644
> --- a/lib/stackdepot.c
> +++ b/lib/stackdepot.c
> @@ -681,6 +681,14 @@ depot_stack_handle_t stack_depot_save(unsigned long *entries,
>  }
>  EXPORT_SYMBOL_GPL(stack_depot_save);
>
> +struct stack_record *stack_depot_get_stack(depot_stack_handle_t handle)
> +{
> +       if (!handle)
> +               return NULL;
> +
> +       return depot_fetch_stack(handle);
> +}
> +
>  unsigned int stack_depot_fetch(depot_stack_handle_t handle,
>                                unsigned long **entries)
>  {
> diff --git a/mm/page_owner.c b/mm/page_owner.c
> index 5634e5d890f8..0adf41702b9d 100644
> --- a/mm/page_owner.c
> +++ b/mm/page_owner.c
> @@ -61,6 +61,22 @@ static __init bool need_page_owner(void)
>         return page_owner_enabled;
>  }
>
> +static void inc_stack_record_count(depot_stack_handle_t handle)
> +{
> +       struct stack_record *stack = stack_depot_get_stack(handle);
> +
> +       if (stack)
> +               refcount_inc(&stack->count);
> +}
> +
> +static void dec_stack_record_count(depot_stack_handle_t handle)
> +{
> +       struct stack_record *stack = stack_depot_get_stack(handle);
> +
> +       if (stack)
> +               refcount_dec(&stack->count);
> +}
> +
>  static __always_inline depot_stack_handle_t create_dummy_stack(void)
>  {
>         unsigned long entries[4];
> @@ -140,6 +156,7 @@ void __reset_page_owner(struct page *page, unsigned short order)
>         int i;
>         struct page_ext *page_ext;
>         depot_stack_handle_t handle;
> +       depot_stack_handle_t alloc_handle;
>         struct page_owner *page_owner;
>         u64 free_ts_nsec = local_clock();
>
> @@ -147,6 +164,9 @@ void __reset_page_owner(struct page *page, unsigned short order)
>         if (unlikely(!page_ext))
>                 return;
>
> +       page_owner = get_page_owner(page_ext);
> +       alloc_handle = page_owner->handle;
> +
>         handle = save_stack(GFP_NOWAIT | __GFP_NOWARN);
>         for (i = 0; i < (1 << order); i++) {
>                 __clear_bit(PAGE_EXT_OWNER_ALLOCATED, &page_ext->flags);
> @@ -158,6 +178,15 @@ void __reset_page_owner(struct page *page, unsigned short order)
>                 page_ext = page_ext_next(page_ext);
>         }
>         page_ext_put(page_ext);
> +       if (alloc_handle != early_handle)
> +               /*
> +                * early_handle is being set as a handle for all those
> +                * early allocated pages. See init_pages_in_zone().
> +                * Since their refcount is not being incremented because
> +                * the machinery is not ready yet, we cannot decrement
> +                * their refcount either.
> +                */
> +               dec_stack_record_count(alloc_handle);
>  }
>
>  static inline void __set_page_owner_handle(struct page_ext *page_ext,
> @@ -199,6 +228,7 @@ noinline void __set_page_owner(struct page *page, unsigned short order,
>                 return;
>         __set_page_owner_handle(page_ext, handle, order, gfp_mask);
>         page_ext_put(page_ext);
> +       inc_stack_record_count(handle);
>  }
>
>  void __set_page_owner_migrate_reason(struct page *page, int reason)
> --
> 2.43.0
>