[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20210222150233.j5l5u43psfcjgbva@skbuf>
Date: Mon, 22 Feb 2021 17:02:33 +0200
From: Vladimir Oltean <olteanv@...il.com>
To: netdev@...r.kernel.org
Cc: Andrew Lunn <andrew@...n.ch>,
Florian Fainelli <f.fainelli@...il.com>,
Vivien Didelot <vivien.didelot@...il.com>,
DENG Qingfang <dqfext@...il.com>,
Tobias Waldekranz <tobias@...dekranz.com>,
George McCollister <george.mccollister@...il.com>,
Horatiu Vultur <horatiu.vultur@...rochip.com>,
Kurt Kanzenbach <kurt@...utronix.de>,
Marek Behun <marek.behun@....cz>
Subject: Re: [RFC PATCH net-next] selftests: net: dsa: add a test for ports
matching on notifiers
On Mon, Feb 22, 2021 at 02:02:48PM +0200, Vladimir Oltean wrote:
> +/* CPU
> + * |
> + * sw0p0 sw0p1 sw0p2 sw0p3 sw0p4
> + * | DSA link
> + * sw1p0 sw1p1 sw1p2 sw1p3 sw1p4
> + * | DSA link
> + * sw2p0 sw2p1 sw2p2 sw2p3 sw2p4
> + */
Ha, that's one idealistic view of reality, where sw1p4 is connected to
two switches at the same time...
I'm still working on this, I'm using it to review and improve the
notifiers that we have today. I just posted it to get some comments
regarding the best way in which we should present this development tool.
Should we add all notifiers from net/dsa/switch.c or should we leave
just this "test" one, and the developer is responsible for copy-pasting
the code they want to test into this C program?
Powered by blists - more mailing lists