netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jesper Dangaard Brouer <jbrouer@redhat.com>
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>
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>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH RFC net-next v2 2/7] net: page_pool: place frag_* fields in one cacheline
Date: Fri, 14 Jul 2023 20:37:39 +0200	[thread overview]
Message-ID: <ac97825d-6a27-f121-4cee-9d2ee0934ce6@redhat.com> (raw)
In-Reply-To: <20230714170853.866018-3-aleksander.lobakin@intel.com>



On 14/07/2023 19.08, Alexander Lobakin wrote:
> On x86_64, frag_* fields of struct page_pool are scattered across two
> cachelines despite the summary size of 24 bytes. The last field,
> ::frag_users, is pushed out to the next one, sharing it with
> ::alloc_stats.
> All three fields are used in pretty much the same places. There are some
> holes and cold members to move around. Move frag_* one block up, placing
> them right after &page_pool_params perfectly at the beginning of CL2.
> This doesn't do any meaningful to the second block, as those are some
> destroy-path cold structures, and doesn't do anything to ::alloc_stats,
> which still starts at 200-byte offset, 8 bytes after CL3 (still fitting
> into 1 cacheline).
> On my setup, this yields 1-2% of Mpps when using PP frags actively.
> When it comes to 32-bit architectures with 32-byte CL: &page_pool_params
> plus ::pad is 44 bytes, the block taken care of is 16 bytes within one
> CL, so there should be at least no regressions from the actual change.
> 
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>   include/net/page_pool.h | 10 +++++-----
>   1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/include/net/page_pool.h b/include/net/page_pool.h
> index 829dc1f8ba6b..212d72b5cfec 100644
> --- a/include/net/page_pool.h
> +++ b/include/net/page_pool.h
> @@ -130,16 +130,16 @@ static inline u64 *page_pool_ethtool_stats_get(u64 *data, void *stats)
>   struct page_pool {
>   	struct page_pool_params p;
>   
> +	long frag_users;
> +	struct page *frag_page;
> +	unsigned int frag_offset;
> +	u32 pages_state_hold_cnt;

I think this is okay, but I want to highlight that:
  - pages_state_hold_cnt and pages_state_release_cnt
need to be kept on separate cache-lines.


> +
>   	struct delayed_work release_dw;
>   	void (*disconnect)(void *);
>   	unsigned long defer_start;
>   	unsigned long defer_warn;
>   
> -	u32 pages_state_hold_cnt;
> -	unsigned int frag_offset;
> -	struct page *frag_page;
> -	long frag_users;
> -
>   #ifdef CONFIG_PAGE_POOL_STATS
>   	/* these stats are incremented while in softirq context */
>   	struct page_pool_alloc_stats alloc_stats;


  parent reply	other threads:[~2023-07-14 18:37 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-14 17:08 [PATCH RFC net-next v2 0/7] net: page_pool: a couple of assorted optimizations Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 1/7] net: skbuff: don't include <net/page_pool.h> to <linux/skbuff.h> Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 2/7] net: page_pool: place frag_* fields in one cacheline Alexander Lobakin
2023-07-14 17:14   ` Alexander Lobakin
2023-07-14 18:37   ` Jesper Dangaard Brouer [this message]
2023-07-18 13:50     ` Alexander Lobakin
2023-07-26  8:13       ` Ilias Apalodimas
2023-07-26 10:39         ` Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 2/7] net: page_pool: shrink &page_pool_params a tiny bit Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 3/7] net: page_pool: place frag_* fields in one cacheline Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 3/7] net: page_pool: shrink &page_pool_params a tiny bit Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 4/7] net: page_pool: don't use driver-set flags field directly Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 5/7] net: page_pool: avoid calling no-op externals when possible Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 6/7] net: skbuff: avoid accessing page_pool if !napi_safe when returning page Alexander Lobakin
2023-07-14 17:08 ` [PATCH RFC net-next v2 7/7] net: skbuff: always try to recycle PP pages directly when in softirq Alexander Lobakin
2023-07-19  0:40   ` Jakub Kicinski
2023-07-19 16:34     ` Alexander Lobakin
2023-07-19 20:51       ` Jakub Kicinski
2023-07-20 16:46         ` Alexander Lobakin
2023-07-20 17:12           ` Jakub Kicinski
2023-07-20 17:48             ` Alexander Lobakin
2023-07-20 18:00               ` Jakub Kicinski
2023-07-20 18:01                 ` Alexander Lobakin
2023-07-20 18:13                   ` Alexander Lobakin
2023-07-20 19:20                     ` Jakub Kicinski
2023-07-20 19:33                       ` Alexander Lobakin
2023-07-20 19:46                         ` Jakub Kicinski
2023-07-21 11:53                           ` Yunsheng Lin
2023-07-21 15:37                             ` Alexander Lobakin
2023-07-21 16:01                               ` Jakub Kicinski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ac97825d-6a27-f121-4cee-9d2ee0934ce6@redhat.com \
    --to=jbrouer@redhat.com \
    --cc=aleksander.lobakin@intel.com \
    --cc=alexanderduyck@fb.com \
    --cc=brouer@redhat.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=hawk@kernel.org \
    --cc=ilias.apalodimas@linaro.org \
    --cc=kuba@kernel.org \
    --cc=larysa.zaremba@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linyunsheng@huawei.com \
    --cc=maciej.fijalkowski@intel.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).