On Thu, Nov 23, 2023 at 2:18 PM Yunsheng Lin <linyunsheng@xxxxxxxxxx> wrote:
>
> On 2023/11/23 10:25, Liang Chen wrote:
> > To support multiple users referencing the same fragment, pp_frag_count is
> > renamed to pp_ref_count to better reflect its actual meaning based on the
> > suggestion from [1].
>
> The renaming looks good to me, some minor nit.
>
> It is good to add a cover-letter using 'git format-patch --cover-letter'
> to explain the overall background or modifications this patchset make when
> there is more than one patch.
>
Thanks for the suggestion. A cover-letter will be provided for the next version.
> >
> > [1]
> > http://lore.kernel.org/netdev/f71d9448-70c8-8793-dc9a-0eb48a570300@xxxxxxxxxx
> >
> > Signed-off-by: Liang Chen <liangchen.linux@xxxxxxxxx>
> > ---
> > include/linux/mm_types.h | 2 +-
> > include/net/page_pool/helpers.h | 31 ++++++++++++++++++-------------
> > 2 files changed, 19 insertions(+), 14 deletions(-)
> >
> > diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
> > index 957ce38768b2..64e4572ef06d 100644
> > --- a/include/linux/mm_types.h
> > +++ b/include/linux/mm_types.h
> > @@ -125,7 +125,7 @@ struct page {
> > struct page_pool *pp;
> > unsigned long _pp_mapping_pad;
> > unsigned long dma_addr;
> > - atomic_long_t pp_frag_count;
> > + atomic_long_t pp_ref_count;
>
> It seems that we may have 4 bytes available for 64 bit arch if we change
> the 'atomic_long_t' to 'refcount_t':)
>
> > };
> > struct { /* Tail pages of compound page */
> > unsigned long compound_head; /* Bit zero is set */
> > diff --git a/include/net/page_pool/helpers.h b/include/net/page_pool/helpers.h
> > index 4ebd544ae977..a6dc9412c9ae 100644
> > --- a/include/net/page_pool/helpers.h
> > +++ b/include/net/page_pool/helpers.h
> > @@ -29,7 +29,7 @@
> > * page allocated from page pool. Page splitting enables memory saving and thus
> > * avoids TLB/cache miss for data access, but there also is some cost to
> > * implement page splitting, mainly some cache line dirtying/bouncing for
> > - * 'struct page' and atomic operation for page->pp_frag_count.
> > + * 'struct page' and atomic operation for page->pp_ref_count.
> > *
> > * The API keeps track of in-flight pages, in order to let API users know when
> > * it is safe to free a page_pool object, the API users must call
> > @@ -214,61 +214,66 @@ inline enum dma_data_direction page_pool_get_dma_dir(struct page_pool *pool)
> > return pool->p.dma_dir;
> > }
> >
> > -/* pp_frag_count represents the number of writers who can update the page
> > +/* pp_ref_count represents the number of writers who can update the page
> > * either by updating skb->data or via DMA mappings for the device.
> > * We can't rely on the page refcnt for that as we don't know who might be
> > * holding page references and we can't reliably destroy or sync DMA mappings
> > * of the fragments.
> > *
> > - * When pp_frag_count reaches 0 we can either recycle the page if the page
> > + * pp_ref_count initially corresponds to the number of fragments. However,
> > + * when multiple users start to reference a single fragment, for example in
> > + * skb_try_coalesce, the pp_ref_count will become greater than the number of
> > + * fragments.
> > + *
> > + * When pp_ref_count reaches 0 we can either recycle the page if the page
> > * refcnt is 1 or return it back to the memory allocator and destroy any
> > * mappings we have.
> > */
> > static inline void page_pool_fragment_page(struct page *page, long nr)
> > {
> > - atomic_long_set(&page->pp_frag_count, nr);
> > + atomic_long_set(&page->pp_ref_count, nr);
> > }
> >
> > static inline long page_pool_defrag_page(struct page *page, long nr)
> > {
> > long ret;
> >
> > - /* If nr == pp_frag_count then we have cleared all remaining
> > + /* If nr == pp_ref_count then we have cleared all remaining
> > * references to the page:
> > * 1. 'n == 1': no need to actually overwrite it.
> > * 2. 'n != 1': overwrite it with one, which is the rare case
> > - * for pp_frag_count draining.
> > + * for pp_ref_count draining.
> > *
> > * The main advantage to doing this is that not only we avoid a atomic
> > * update, as an atomic_read is generally a much cheaper operation than
> > * an atomic update, especially when dealing with a page that may be
> > - * partitioned into only 2 or 3 pieces; but also unify the pp_frag_count
> > + * partitioned into only 2 or 3 pieces; but also unify the pp_ref_count
>
> Maybe "referenced by only 2 or 3 users" is more appropriate now?
>
Sure.
> > * handling by ensuring all pages have partitioned into only 1 piece
> > * initially, and only overwrite it when the page is partitioned into
> > * more than one piece.
> > */
> > - if (atomic_long_read(&page->pp_frag_count) == nr) {
> > + if (atomic_long_read(&page->pp_ref_count) == nr) {
> > /* As we have ensured nr is always one for constant case using
> > * the BUILD_BUG_ON(), only need to handle the non-constant case
> > - * here for pp_frag_count draining, which is a rare case.
> > + * here for pp_ref_count draining, which is a rare case.
> > */
> > BUILD_BUG_ON(__builtin_constant_p(nr) && nr != 1);
> > if (!__builtin_constant_p(nr))
> > - atomic_long_set(&page->pp_frag_count, 1);
> > + atomic_long_set(&page->pp_ref_count, 1);
> >
> > return 0;
> > }
> >
> > - ret = atomic_long_sub_return(nr, &page->pp_frag_count);
> > + ret = atomic_long_sub_return(nr, &page->pp_ref_count);
> > WARN_ON(ret < 0);
> >
> > - /* We are the last user here too, reset pp_frag_count back to 1 to
> > + /* We are the last user here too, reset pp_ref_count back to 1 to
> > * ensure all pages have been partitioned into 1 piece initially,
> > * this should be the rare case when the last two fragment users call
> > * page_pool_defrag_page() currently.
>
> Do we need to rename the page_pool_defrag_page() and page_pool_is_last_frag()
> too?
Yeah, I think so. Once a pp page is drained, its management shifts to being primarily governed by pp_ref_count, and there's no longer a need to consider fragmenting. will be done in the next iteration.
>
> > */
> > if (unlikely(!ret))
> > - atomic_long_set(&page->pp_frag_count, 1);
> > + atomic_long_set(&page->pp_ref_count, 1);
> >
> > return ret;
> > }
> >
