From: Barry Song <21cnbao@gmail.com>
To: Eric Dumazet <edumazet@google.com>
Cc: corbet@lwn.net, davem@davemloft.net, hannes@cmpxchg.org,
horms@kernel.org, jackmanb@google.com, kuba@kernel.org,
kuniyu@google.com, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
linyunsheng@huawei.com, mhocko@suse.com, netdev@vger.kernel.org,
pabeni@redhat.com, surenb@google.com, v-songbaohua@oppo.com,
vbabka@suse.cz, willemb@google.com, zhouhuacai@oppo.com,
ziy@nvidia.com
Subject: Re: [RFC PATCH] mm: net: disable kswapd for high-order network buffer allocation
Date: Tue, 14 Oct 2025 14:43:13 +0800 [thread overview]
Message-ID: <CAGsJ_4ySSn6B+x+4zE0Ld1+AM4q-WnS0LfxzWw22oXr7n5NZ=g@mail.gmail.com> (raw)
In-Reply-To: <CANn89iKCZyYi+J=5t2sdmvtERnknkwXrGi4QRzM9btYUywkDfw@mail.gmail.com>
> >
> > A problem with the existing sysctl is that it only covers the TX path;
> > for the RX path, we also observe that kswapd consumes significant power.
> > I could add the patch below to make it support the RX path, but it feels
> > like a bit of a layer violation, since the RX path code resides in mm
> > and is intended to serve generic users rather than networking, even
> > though the current callers are primarily network-related.
>
> You might have a buggy driver.
We are observing the RX path as follows:
do_softirq
taskset_hi_action
kalPacketAlloc
__netdev_alloc_skb
page_frag_alloc_align
__page_frag_cache_refill
This appears to be a fairly common stack.
So it is a buggy driver?
>
> High performance drivers use order-0 allocations only.
>
Do you have an example of high-performance drivers that use only order-0 memory?
Thanks
Barry
next prev parent reply other threads:[~2025-10-14 6:43 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-13 10:16 Barry Song
2025-10-13 18:30 ` Vlastimil Babka
2025-10-13 21:35 ` Shakeel Butt
2025-10-13 21:53 ` Alexei Starovoitov
2025-10-13 22:25 ` Shakeel Butt
2025-10-13 22:46 ` Roman Gushchin
2025-10-14 4:31 ` Barry Song
2025-10-14 7:24 ` Michal Hocko
2025-10-14 7:26 ` Michal Hocko
2025-10-14 8:08 ` Barry Song
2025-10-14 14:27 ` Shakeel Butt
2025-10-14 15:14 ` Michal Hocko
2025-10-14 17:22 ` Shakeel Butt
2025-10-15 6:21 ` Michal Hocko
2025-10-15 18:26 ` Shakeel Butt
2025-10-13 18:53 ` Eric Dumazet
2025-10-14 3:58 ` Barry Song
2025-10-14 5:07 ` Eric Dumazet
2025-10-14 6:43 ` Barry Song [this message]
2025-10-14 7:01 ` Eric Dumazet
2025-10-14 8:17 ` Barry Song
2025-10-14 8:25 ` Eric Dumazet
2025-10-13 21:56 ` Matthew Wilcox
2025-10-14 4:09 ` Barry Song
2025-10-14 5:04 ` Eric Dumazet
2025-10-14 8:58 ` Barry Song
2025-10-14 9:49 ` Eric Dumazet
2025-10-14 10:19 ` Barry Song
2025-10-14 10:39 ` Eric Dumazet
2025-10-14 20:17 ` Barry Song
2025-10-15 6:39 ` Eric Dumazet
2025-10-15 7:35 ` Barry Song
2025-10-15 16:39 ` Suren Baghdasaryan
2025-10-14 14:37 ` Shakeel Butt
2025-10-14 20:28 ` Barry Song
2025-10-15 18:13 ` Shakeel Butt
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='CAGsJ_4ySSn6B+x+4zE0Ld1+AM4q-WnS0LfxzWw22oXr7n5NZ=g@mail.gmail.com' \
--to=21cnbao@gmail.com \
--cc=corbet@lwn.net \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=hannes@cmpxchg.org \
--cc=horms@kernel.org \
--cc=jackmanb@google.com \
--cc=kuba@kernel.org \
--cc=kuniyu@google.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linyunsheng@huawei.com \
--cc=mhocko@suse.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=surenb@google.com \
--cc=v-songbaohua@oppo.com \
--cc=vbabka@suse.cz \
--cc=willemb@google.com \
--cc=zhouhuacai@oppo.com \
--cc=ziy@nvidia.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