netdev - Re: [PATCH net-next v4 1/5] net: hdlc_fr: Simpify fr_rx by using "goto rx

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: <CAJht_EMdbGQdXhYJ7xa_R-j-73fbsEjSUeavov40W52aGvQ21g@mail.gmail.com>
Date:   Fri, 30 Oct 2020 13:01:45 -0700
From:   Xie He <xie.he.0141@...il.com>
To:     Willem de Bruijn <willemdebruijn.kernel@...il.com>
Cc:     Jakub Kicinski <kuba@...nel.org>,
        "David S. Miller" <davem@...emloft.net>,
        Network Development <netdev@...r.kernel.org>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Krzysztof Halasa <khc@...waw.pl>
Subject: Re: [PATCH net-next v4 1/5] net: hdlc_fr: Simpify fr_rx by using
 "goto rx_drop" to drop frames

On Fri, Oct 30, 2020 at 9:35 AM Willem de Bruijn
<willemdebruijn.kernel@...il.com> wrote:
>
> In general we try to avoid changing counter behavior like that, as
> existing users
> may depend on current behavior, e.g., in dashboards or automated monitoring.
>
> I don't know how realistic that is in this specific case, no strong
> objections. Use
> good judgment.

Originally this function only increases stats.rx_dropped only when
there's a memory squeeze. I don't know the specification for the
meaning of stats.rx_dropped, but as I understand it indicates a frame
is dropped. This is why I wanted to increase it whenever we drop a
frame.

Originally this function drops a frame silently if the PVC virtual
device that corresponds to the DLCI number and the protocol type
doesn't exist. I think we may at least need some way to note this.
Originally this function drops a frame with a kernel info message
printed if the protocol type is not supported. I think this is a bad
way because if the other end continuously sends us a lot of frames
with unsupported protocol types, our kernel message log will be
overwhelmed.

I don't know how important it is to keep backwards compatibility. I
usually don't consider this too much. But I can drop this change if we
really want to keep the counter behavior unchanged. I think changing
it is better if we don't consider backwards compatibility.