On Mon, Feb 13, 2017 at 1:09 PM, Eric Dumazet <eduma...@google.com> wrote:
> On Mon, Feb 13, 2017 at 12:51 PM, Alexander Duyck
> <alexander.du...@gmail.com> wrote:
>> On Mon, Feb 13, 2017 at 11:58 AM, Eric Dumazet <eduma...@google.com> wrote:
>
>>> + PAGE_SIZE, priv->dma_dir);
>>> page = page_alloc->page;
>>> /* Revert changes done by mlx4_alloc_pages */
>>> - page_ref_sub(page, page_alloc->page_size /
>>> + page_ref_sub(page, PAGE_SIZE /
>>> priv->frag_info[i].frag_stride - 1);
>>> put_page(page);
>>> page_alloc->page = NULL;
>>
>> You can probably simplify this by using __page_frag_cache_drain since
>> that way you can just doe the sub and test for the whole set instead
>> of doing N - 1 and 1.
>
> Well, we remove this anyway in following patch ;)
Any plans to add the bulk page count updates back at some point? I
just got around to adding it for igb in commit bd4171a5d4c2 ("igb:
update code to better handle incrementing page count"). I should have
patches for ixgbe, i40e, and i40evf here pretty soon. That was one of
the reasons why I implemented __page_frag_cache_drain in the first
place. As I'm sure Jesper will probably point out the atomic op for
get_page/page_ref_inc can be pretty expensive if I recall correctly.
- Alex
