[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20170622.130528.1762873686654379973.davem@davemloft.net>
Date: Thu, 22 Jun 2017 13:05:28 -0400 (EDT)
From: David Miller <davem@...emloft.net>
To: willemdebruijn.kernel@...il.com
Cc: netdev@...r.kernel.org, linux-api@...r.kernel.org,
willemb@...gle.com
Subject: Re: [PATCH net-next v3 02/13] sock: skb_copy_ubufs support for
compound pages
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
Date: Wed, 21 Jun 2017 17:18:05 -0400
> @@ -958,15 +958,20 @@ EXPORT_SYMBOL_GPL(skb_morph);
> */
> int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask)
> {
> - int i;
> int num_frags = skb_shinfo(skb)->nr_frags;
> struct page *page, *head = NULL;
> struct ubuf_info *uarg = skb_shinfo(skb)->destructor_arg;
> + int i, new_frags;
> + u32 d_off;
Reverse christmas tree with the local variable declarations.
> + page = head;
> + d_off = 0;
> + for (i = 0; i < num_frags; i++) {
> + u8 *vaddr;
> + skb_frag_t *f = &skb_shinfo(skb)->frags[i];
> + u32 f_off, f_size, copy;
Likewise.
> + f_off = f->page_offset;
> + f_size = f->size;
> +
> + vaddr = kmap_atomic(skb_frag_page(f));
I looked at some kmap_atomic() implementations and I do not think
it supports compound pages.
Powered by blists - more mailing lists