netdev - Re: [PATCH RFC net-next/mm V1 2/3] page_pool: Use static

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20230425233027.w3olphld4nkcdvry@dhcp-172-26-102-232.dhcp.thefacebook.com>
Date:   Tue, 25 Apr 2023 16:30:27 -0700
From:   Alexei Starovoitov <alexei.starovoitov@...il.com>
To:     Jesper Dangaard Brouer <brouer@...hat.com>
Cc:     Ilias Apalodimas <ilias.apalodimas@...aro.org>,
        netdev@...r.kernel.org, Eric Dumazet <eric.dumazet@...il.com>,
        linux-mm@...ck.org, Mel Gorman <mgorman@...hsingularity.net>,
        lorenzo@...nel.org,
        Toke Høiland-Jørgensen <toke@...hat.com>,
        linyunsheng@...wei.com, bpf@...r.kernel.org,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Andrew Morton <akpm@...ux-foundation.org>, willy@...radead.org
Subject: Re: [PATCH RFC net-next/mm V1 2/3] page_pool: Use static_key for
 shutdown phase

On Tue, Apr 25, 2023 at 07:15:43PM +0200, Jesper Dangaard Brouer wrote:
> Performance is very important for page pool (PP). This add the use of
> static_key APIs for regaining a single instruction, which makes the
> new PP shutdown scheme zero impact.
> 
> We are uncertain if this is 100% correct, because static_key APIs uses
> a mutex lock and it is uncertain if all contexts that can return pages
> can support this. We could spawn a workqueue (like we just removed) to
> workaround this issue.

With debug atomic sleep the issue should be trivial to see.
iirc the callers of xdp_flush_frame_bulk() need to do it under rcu_read_lock equivalent,
which is not sleepable and mutex-es should warn.