Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Jesper Dangaard Brouer <jbrouer@redhat.com>
Message-ID: <db85d260-fdad-9b7c-cf7e-2e848151292d@redhat.com>
Date:   Fri, 28 Jul 2023 11:32:01 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.13.0
Cc:     brouer@redhat.com,
        Maciej Fijalkowski <maciej.fijalkowski@intel.com>,
        Larysa Zaremba <larysa.zaremba@intel.com>,
        Yunsheng Lin <linyunsheng@huawei.com>,
        Alexander Duyck <alexanderduyck@fb.com>,
        Jesper Dangaard Brouer <hawk@kernel.org>,
        Ilias Apalodimas <ilias.apalodimas@linaro.org>,
        Simon Horman <simon.horman@corigine.com>,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
        qingfang.deng@siflower.com.cn
Subject: Re: [PATCH net-next 9/9] net: skbuff: always try to recycle PP pages
 directly when in softirq
Content-Language: en-US
To:     Alexander Lobakin <aleksander.lobakin@intel.com>,
        "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>
References: <20230727144336.1646454-1-aleksander.lobakin@intel.com>
 <20230727144336.1646454-10-aleksander.lobakin@intel.com>
In-Reply-To: <20230727144336.1646454-10-aleksander.lobakin@intel.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 27/07/2023 16.43, Alexander Lobakin wrote:
> Commit 8c48eea3adf3 ("page_pool: allow caching from safely localized
> NAPI") allowed direct recycling of skb pages to their PP for some cases,
> but unfortunately missed a couple of other majors.
> For example, %XDP_DROP in skb mode. The netstack just calls kfree_skb(),
> which unconditionally passes `false` as @napi_safe. Thus, all pages go
> through ptr_ring and locks, although most of time we're actually inside
> the NAPI polling this PP is linked with, so that it would be perfectly
> safe to recycle pages directly.

The commit messages is hard to read. It would help me as the reader if
you used a empty line between paragraphs, like in this location (same
goes for other commit descs).

> Let's address such. If @napi_safe is true, we're fine, don't change
> anything for this path. But if it's false, check whether we are in the
> softirq context. It will most likely be so and then if ->list_owner
> is our current CPU, we're good to use direct recycling, even though
> @napi_safe is false -- concurrent access is excluded. in_softirq()
> protection is needed mostly due to we can hit this place in the
> process context (not the hardirq though).

This patch make me a little nervous, as it can create hard-to-debug bugs
if this isn't 100% correct.  (Thanks for previous patch that exclude
hardirq via lockdep).

> For the mentioned xdp-drop-skb-mode case, the improvement I got is
> 3-4% in Mpps. As for page_pool stats, recycle_ring is now 0 and
> alloc_slow counter doesn't change most of time, which means the
> MM layer is not even called to allocate any new pages.
> 
> Suggested-by: Jakub Kicinski <kuba@kernel.org> # in_softirq()
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>   net/core/skbuff.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
> index e701401092d7..5ba3948cceed 100644
> --- a/net/core/skbuff.c
> +++ b/net/core/skbuff.c
> @@ -901,8 +901,10 @@ bool page_pool_return_skb_page(struct page *page, bool napi_safe)
>   	/* 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.
> +	 * __page_pool_put_page() makes sure we're not in hardirq context
> +	 * and interrupts are enabled prior to accessing the cache.
>   	 */
> -	if (napi_safe) {
> +	if (napi_safe || in_softirq()) {

I used to have in_serving_softirq() in PP to exclude process context
that just disabled BH to do direct recycling (into a lockless array).
This changed in kernel v6.3 commit 542bcea4be86 ("net: page_pool: use
in_softirq() instead") to help threaded NAPI.  I guess, nothing blew up
so I guess this was okay to relax this.

>   		const struct napi_struct *napi = READ_ONCE(pp->p.napi);
>   
>   		allow_direct = napi &&

AFAIK this in_softirq() will allow process context with disabled BH to
also recycle directly into the PP lockless array.  With the additional
checks (that are just outside above diff-context) that I assume makes
sure CPU (smp_processor_id()) also match.  Is this safe?

--Jesper