From: Joe Damato <jdamato@fastly.com>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: linux-rdma@vger.kernel.org, netdev@vger.kernel.org,
"David S. Miller" <davem@davemloft.net>,
Andrew Lunn <andrew+netdev@lunn.ch>,
Eric Dumazet <edumazet@google.com>,
Ilias Apalodimas <ilias.apalodimas@linaro.org>,
Jakub Kicinski <kuba@kernel.org>,
Jesper Dangaard Brouer <hawk@kernel.org>,
Leon Romanovsky <leon@kernel.org>,
Paolo Abeni <pabeni@redhat.com>,
Saeed Mahameed <saeedm@nvidia.com>,
Simon Horman <horms@kernel.org>, Tariq Toukan <tariqt@nvidia.com>,
Thomas Gleixner <tglx@linutronix.de>,
Yunsheng Lin <linyunsheng@huawei.com>
Subject: Re: [PATCH net-next 1/2] page_pool: Convert page_pool_recycle_stats to u64_stats_t.
Date: Fri, 21 Feb 2025 12:21:40 -0500 [thread overview]
Message-ID: <Z7i2JHiKX6rggsUz@LQ3V64L9R2> (raw)
In-Reply-To: <20250221115221.291006-2-bigeasy@linutronix.de>
On Fri, Feb 21, 2025 at 12:52:20PM +0100, Sebastian Andrzej Siewior wrote:
> Using u64 for statistics can lead to inconsistency on 32bit because an
> update and a read requires to access two 32bit values.
> This can be avoided by using u64_stats_t for the counters and
> u64_stats_sync for the required synchronisation on 32bit platforms. The
> synchronisation is a NOP on 64bit architectures.
As mentioned in my response to the cover letter, I'd want to see
before/after 32bit assembly to ensure that this assertion is
correct.
[...]
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
> index 611ec4b6f3709..baff961970f25 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
> @@ -501,7 +501,7 @@ static void mlx5e_stats_update_stats_rq_page_pool(struct mlx5e_channel *c)
> {
> struct mlx5e_rq_stats *rq_stats = c->rq.stats;
> struct page_pool *pool = c->rq.page_pool;
> - struct page_pool_stats stats = { 0 };
> + struct page_pool_stats stats = { };
>
> if (!page_pool_get_stats(pool, &stats))
> return;
> @@ -513,11 +513,11 @@ static void mlx5e_stats_update_stats_rq_page_pool(struct mlx5e_channel *c)
> rq_stats->pp_alloc_waive = stats.alloc_stats.waive;
> rq_stats->pp_alloc_refill = stats.alloc_stats.refill;
>
> - rq_stats->pp_recycle_cached = stats.recycle_stats.cached;
> - rq_stats->pp_recycle_cache_full = stats.recycle_stats.cache_full;
> - rq_stats->pp_recycle_ring = stats.recycle_stats.ring;
> - rq_stats->pp_recycle_ring_full = stats.recycle_stats.ring_full;
> - rq_stats->pp_recycle_released_ref = stats.recycle_stats.released_refcnt;
> + rq_stats->pp_recycle_cached = u64_stats_read(&stats.recycle_stats.cached);
> + rq_stats->pp_recycle_cache_full = u64_stats_read(&stats.recycle_stats.cache_full);
> + rq_stats->pp_recycle_ring = u64_stats_read(&stats.recycle_stats.ring);
> + rq_stats->pp_recycle_ring_full = u64_stats_read(&stats.recycle_stats.ring_full);
> + rq_stats->pp_recycle_released_ref = u64_stats_read(&stats.recycle_stats.released_refcnt);
> }
> #else
> static void mlx5e_stats_update_stats_rq_page_pool(struct mlx5e_channel *c)
It might be better to convert mlx5 to
page_pool_ethtool_stats_get_strings and
page_pool_ethtool_stats_get_count instead ?
next prev parent reply other threads:[~2025-02-21 17:21 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-21 11:52 [PATCH net-next 0/2] page_pool: Convert stats to u64_stats_t Sebastian Andrzej Siewior
2025-02-21 11:52 ` [PATCH net-next 1/2] page_pool: Convert page_pool_recycle_stats " Sebastian Andrzej Siewior
2025-02-21 17:21 ` Joe Damato [this message]
2025-02-26 12:06 ` Sebastian Andrzej Siewior
2025-02-22 8:13 ` Yunsheng Lin
2025-02-25 11:27 ` Paolo Abeni
2025-02-26 9:28 ` Sebastian Andrzej Siewior
2025-02-21 11:52 ` [PATCH net-next 2/2] page_pool: Convert page_pool_alloc_stats " Sebastian Andrzej Siewior
2025-02-21 17:30 ` Joe Damato
2025-02-22 8:13 ` Yunsheng Lin
2025-02-21 17:10 ` [PATCH net-next 0/2] page_pool: Convert stats " Joe Damato
2025-02-26 10:27 ` Sebastian Andrzej Siewior
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=Z7i2JHiKX6rggsUz@LQ3V64L9R2 \
--to=jdamato@fastly.com \
--cc=andrew+netdev@lunn.ch \
--cc=bigeasy@linutronix.de \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=hawk@kernel.org \
--cc=horms@kernel.org \
--cc=ilias.apalodimas@linaro.org \
--cc=kuba@kernel.org \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=linyunsheng@huawei.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=saeedm@nvidia.com \
--cc=tariqt@nvidia.com \
--cc=tglx@linutronix.de \
/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