[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250409161558.8a8WSoml@linutronix.de>
Date: Wed, 9 Apr 2025 18:15:58 +0200
From: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To: Jakub Kicinski <kuba@...nel.org>
Cc: linux-rt-devel@...ts.linux.dev, netdev@...r.kernel.org,
"David S. Miller" <davem@...emloft.net>,
Andrew Lunn <andrew+netdev@...n.ch>,
Eric Dumazet <edumazet@...gle.com>,
Ilias Apalodimas <ilias.apalodimas@...aro.org>,
Jesper Dangaard Brouer <hawk@...nel.org>,
Joe Damato <jdamato@...tly.com>, Paolo Abeni <pabeni@...hat.com>,
Saeed Mahameed <saeedm@...dia.com>, Simon Horman <horms@...nel.org>,
Tariq Toukan <tariqt@...dia.com>,
Thomas Gleixner <tglx@...utronix.de>,
Yunsheng Lin <linyunsheng@...wei.com>
Subject: Re: [PATCH net-next v3 0/4] page_pool: Convert stats to u64_stats_t.
On 2025-04-08 18:56:36 [-0700], Jakub Kicinski wrote:
> On Tue, 8 Apr 2025 12:59:17 +0200 Sebastian Andrzej Siewior wrote:
> > I don't know if it is ensured that only *one* update can happen because
> > the stats are per-CPU and per NAPI device. But there will be now a
> > warning on 32bit if this is really attempted in preemptible context.
>
> I think recycling may happen in preemptible context, and from BH.
> Have you tried to test?
Let me try to find something for testing. My mlx box has no ethernet
cable plugged.
Yunsheng mentioned also something that there might be more than one
writer this might ask for different approach.
> The net.core.skb_defer_max sysctl may mask it for TCP traffic.
Sebastian
Powered by blists - more mailing lists