[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <61965a16-55bf-2f81-8938-856f834d35bf@intel.com>
Date: Tue, 1 Aug 2023 15:25:38 +0200
From: Alexander Lobakin <aleksander.lobakin@...el.com>
To: Yunsheng Lin <linyunsheng@...wei.com>
CC: "David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
Maciej Fijalkowski <maciej.fijalkowski@...el.com>,
Larysa Zaremba <larysa.zaremba@...el.com>,
Alexander Duyck <alexanderduyck@...com>,
Jesper Dangaard Brouer <hawk@...nel.org>,
"Ilias Apalodimas" <ilias.apalodimas@...aro.org>,
Simon Horman <simon.horman@...igine.com>,
<netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH net-next 2/9] net: skbuff: don't include
<net/page_pool/types.h> to <linux/skbuff.h>
From: Yunsheng Lin <linyunsheng@...wei.com>
Date: Sat, 29 Jul 2023 19:40:19 +0800
> On 2023/7/28 21:58, Alexander Lobakin wrote:
>> From: Yunsheng Lin <linyunsheng@...wei.com>
>> Date: Fri, 28 Jul 2023 20:02:51 +0800
>>
>>> On 2023/7/27 22:43, Alexander Lobakin wrote:
>>>> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
>>>
>>> ...
>>>
>>>> +bool page_pool_return_skb_page(struct page *page, bool napi_safe)
>>>
>>> Still having the 'page_pool_' prefix seems odd here when it is in the
>>> skbuff.c where most have skb_ or napi_ prefix, is it better to rename
>>> it to something like napi_return_page_pool_page()?
>>
>> Given that how the function that goes next is named, maybe
>> skb_pp_return_page() (or skb_pp_put_page())?
>
> skb_pp_put_page() seems better.
>
> And I like napi_pp_put_page() with 'napi_' prefix better as
> it does not take a skb as parameter and the naming is aligned
> with the 'napi_safe' parameter.
Ah, I see. Sounds reasonable. I'll pick napi_pp_put_page() for the next
round, I like this one.
>
>>
>>>
>>>> +{
>>>> + struct napi_struct *napi;
>>>> + struct page_pool *pp;
>>>> + bool allow_direct;
>>>> +
>>>> + page = compound_head(page);
>>>> +
>>>> + /* page->pp_magic is OR'ed with PP_SIGNATURE after the allocation
>>>> + * in order to preserve any existing bits, such as bit 0 for the
>>>> + * head page of compound page and bit 1 for pfmemalloc page, so
>>>> + * mask those bits for freeing side when doing below checking,
>>>> + * and page_is_pfmemalloc() is checked in __page_pool_put_page()
>>>> + * to avoid recycling the pfmemalloc page.
>>>> + */
>>>> + if (unlikely((page->pp_magic & ~0x3UL) != PP_SIGNATURE))
>>>> + return false;
>>>> +
>>>> + pp = page->pp;
>>>> +
>>>> + /* Allow direct recycle if we have reasons to believe that we are
>>>> + * in the same context as the consumer would run, so there's
>>>> + * no possible race.
>>>> + */
>>>> + napi = READ_ONCE(pp->p.napi);
>>>> + allow_direct = napi_safe && napi &&
>>>> + READ_ONCE(napi->list_owner) == smp_processor_id();
>>>> +
>>>> + /* Driver set this to memory recycling info. Reset it on recycle.
>>>> + * This will *not* work for NIC using a split-page memory model.
>>>> + * The page will be returned to the pool here regardless of the
>>>> + * 'flipped' fragment being in use or not.
>>>> + */
>>>> + page_pool_put_full_page(pp, page, allow_direct);
>>>> +
>>>> + return true;
>>>> +}
>>>> +EXPORT_SYMBOL(page_pool_return_skb_page);
>>>> +
>>>> static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe)
>>
>> (this one)
>>
>>>> {
>>>> if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)
>
> We may need the 'IS_ENABLED(CONFIG_PAGE_POOL' checking in the newly
> moved function too.
The first person who noticed this, for sure we should have it!
>
>>>>
>>
>> Thanks,
>> Olek
>>
>> .
>>
Thanks,
Olek
Powered by blists - more mailing lists