netdev - Re: [PATCH net v2 0/3] Add missing xdp_do

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

Message-ID: <cb2f7931-5ae5-8583-acff-4a186fed6632@kernel.org>
Date: Wed, 20 Sep 2023 09:04:27 +0200
From: Jesper Dangaard Brouer <hawk@...nel.org>
To: Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
 netdev@...r.kernel.org, bpf@...r.kernel.org
Cc: "David S. Miller" <davem@...emloft.net>,
 Alexei Starovoitov <ast@...nel.org>, Daniel Borkmann <daniel@...earbox.net>,
 Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>,
 Jesper Dangaard Brouer <hawk@...nel.org>,
 John Fastabend <john.fastabend@...il.com>, Paolo Abeni <pabeni@...hat.com>,
 Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH net v2 0/3] Add missing xdp_do_flush() invocations.

Hi Sebastian,

On 18/09/2023 17.36, Sebastian Andrzej Siewior wrote:
> Hi,
> 
> I've been looking at the drivers/ XDP users and noticed that some
> XDP_REDIRECT user don't invoke xdp_do_flush() at the end.

I'm wondering if we could detect (and WARN) in the net core e.g.
net_rx_action() that a driver is missing a flush?

The idea could be to check the per CPU (struct) bpf_redirect_info.
Or check (per CPU) dev_flush_list.

If some is worried about performance implications, then we can hide this
under CONFIG_DEBUG_NET.

> v1…v2: https://lore.kernel.org/all/20230908135748.794163-1-bigeasy@linutronix.de
>    - Collected tags.
>    - Dropped the #4 patch which was touching cpu_map_bpf_prog_run()
>      because it is not needed.
> 

For the fixes in this set:

Acked-by: Jesper Dangaard Brouer <hawk@...nel.org>

Thanks for fixing these! - at it can lead to strange to debug cases.
--Jesper