On Mon, 17 Feb 2020 09:46:08 +0200 Ilias Apalodimas <ilias.apalodimas@xxxxxxxxxx> wrote: > On Mon, Feb 17, 2020 at 08:41:33AM +0100, Jesper Dangaard Brouer wrote: > > On Mon, 17 Feb 2020 08:28:49 +0200 > > Ilias Apalodimas <ilias.apalodimas@xxxxxxxxxx> wrote: > > > > > diff --git a/include/net/page_pool.h b/include/net/page_pool.h > > > index cfbed00ba7ee..7c1f23930035 100644 > > > --- a/include/net/page_pool.h > > > +++ b/include/net/page_pool.h > > > @@ -162,39 +162,33 @@ static inline void page_pool_use_xdp_mem(struct page_pool *pool, > > > } > > > #endif > > > > > > -/* Never call this directly, use helpers below */ > > > -void __page_pool_put_page(struct page_pool *pool, struct page *page, > > > - unsigned int dma_sync_size, bool allow_direct); > > > +void page_pool_release_page(struct page_pool *pool, struct page *page); > > > > > > -static inline void page_pool_put_page(struct page_pool *pool, > > > - struct page *page, bool allow_direct) > > > +/* If the page refcnt == 1, this will try to recycle the page. > > > + * if PP_FLAG_DMA_SYNC_DEV is set, it will try to sync the DMA area for > > > + * the configured size min(dma_sync_size, pool->max_len). > > > + * If the page refcnt != page will be returned > > > > Is this last comment line fully formed? > > Yes, but that dosen't mena it makes sense! > Maybe i should switch the last sentence to sometning like: > "If the page refcnt != 1, page will be returned to memory subsystem" ? Yes, that sounds better. -- Best regards, Jesper Dangaard Brouer MSc.CS, Principal Kernel Engineer at Red Hat LinkedIn: http://www.linkedin.com/in/brouer
