netdev - Re: [PATCH net-next 3/8] net: enqueue_to_backlog() change vs not running device

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: <CANn89iLpY7iWppW1Vxze6Gf0ki5YFN9qF-w=+ig+=YfLqaLZyg@mail.gmail.com>
Date: Fri, 29 Mar 2024 07:31:19 +0100
From: Eric Dumazet <edumazet@...gle.com>
To: Jason Xing <kerneljasonxing@...il.com>
Cc: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, 
	Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org, eric.dumazet@...il.com
Subject: Re: [PATCH net-next 3/8] net: enqueue_to_backlog() change vs not
 running device

On Fri, Mar 29, 2024 at 4:21 AM Jason Xing <kerneljasonxing@...il.com> wrote:
>
> On Fri, Mar 29, 2024 at 1:07 AM Eric Dumazet <edumazet@...gle.com> wrote:
> >
> > If the device attached to the packet given to enqueue_to_backlog()
> > is not running, we drop the packet.
> >
> > But we accidentally increase sd->dropped, giving false signals
> > to admins: sd->dropped should be reserved to cpu backlog pressure,
> > not to temporary glitches at device dismantles.
>
> It seems that drop action happening is intended in this case (see
> commit e9e4dd3267d0c ("net: do not process device backlog during
> unregistration")). We can see the strange/unexpected behaviour at
> least through simply taking a look at /proc/net/softnet_stat file.

I disagree.

We are dismantling a device, temporary drops are expected, and this
patch adds a more precise drop_reason.

I have seen admins being worried about this counter being not zero on
carefully tuned hosts.

If you think we have to carry these drops forever in
/proc/net/softnet_stat, you will have give
a more precise reason than "This was intentionally added in 2015"

e9e4dd3267d0c5234c changelog was very long, but said nothing
about why sd->dropped _had_ to be updated.