netdev - Re: [RFC net-next 1/2] page_pool: allow caching from safely localized NAPI

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Wed, 5 Apr 2023 19:11:20 +0200
From:   Eric Dumazet <edumazet@...gle.com>
To:     Yunsheng Lin <linyunsheng@...wei.com>
Cc:     Jakub Kicinski <kuba@...nel.org>, davem@...emloft.net,
        netdev@...r.kernel.org, pabeni@...hat.com, hawk@...nel.org,
        ilias.apalodimas@...aro.org
Subject: Re: [RFC net-next 1/2] page_pool: allow caching from safely localized NAPI

On Tue, Apr 4, 2023 at 12:50 PM Yunsheng Lin <linyunsheng@...wei.com> wrote:

> Is there any reason not to introduce back the per socket defer_list to
> not acquire the defer_lock for one skb at a time instead of adding
> per-cpu caches?

This is a disaster on hosts with millions of sockets.
Just think how many pages _one_ skb can hold, and multiply this number
by 10,000,000

per-cpu is the more natural way to make sure this kind of cache will
not use an unbound amount of memory.

We had the same issues with sk->sk_forward_alloc per-socket reserves.