From: Jakub Kicinski <kuba@kernel.org>
To: Nimrod Oren <noren@nvidia.com>
Cc: Jesper Dangaard Brouer <hawk@kernel.org>,
Ilias Apalodimas <ilias.apalodimas@linaro.org>,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Paolo Abeni <pabeni@redhat.com>, Simon Horman <horms@kernel.org>,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
Gal Pressman <gal@nvidia.com>,
Dragos Tatulea <dtatulea@nvidia.com>,
Tariq Toukan <tariqt@nvidia.com>
Subject: Re: [RFC net-next] net: page_pool: cap alloc cache size and refill by pool ring size
Date: Thu, 26 Feb 2026 08:56:43 -0800 [thread overview]
Message-ID: <20260226085643.20b9f256@kernel.org> (raw)
In-Reply-To: <03b14f08-bb37-4130-b81c-6993ff406df0@nvidia.com>
On Thu, 26 Feb 2026 17:28:55 +0200 Nimrod Oren wrote:
> On 25/02/2026 1:39, Jakub Kicinski wrote:
> > I'd simply change the defines based on PAGE_SIZE. The allocation batch
> > size has nothing to do with the ring size, it's just amortizing
> > allocations within a single NAPI cycle.
>
> Thanks, that sounds good to me.
> Do you think a formula like this would work?
>
> #define PP_ALLOC_CACHE_REFILL ((64 * SZ_4K) / PAGE_SIZE)
> #define PP_ALLOC_CACHE_SIZE (PP_ALLOC_CACHE_REFILL * 2)
>
> It keeps both values constant in bytes across page sizes, ensuring a
> consistent memory footprint for pool alloc-caches regardless of a
> system's page size.
I like the clarity / obviousness of the if/elif/else list.
No need to think what the result would be for a given page size.
But up to you.
prev parent reply other threads:[~2026-02-26 16:56 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-23 9:24 [RFC net-next] net: page_pool: cap alloc cache size and refill by pool ring size Nimrod Oren
2026-02-24 23:39 ` Jakub Kicinski
2026-02-26 15:28 ` Nimrod Oren
2026-02-26 16:56 ` Jakub Kicinski [this message]
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=20260226085643.20b9f256@kernel.org \
--to=kuba@kernel.org \
--cc=davem@davemloft.net \
--cc=dtatulea@nvidia.com \
--cc=edumazet@google.com \
--cc=gal@nvidia.com \
--cc=hawk@kernel.org \
--cc=horms@kernel.org \
--cc=ilias.apalodimas@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=noren@nvidia.com \
--cc=pabeni@redhat.com \
--cc=tariqt@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.