[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6196fdbb.1c69fb81.e8741.2b6c@mx.google.com>
Date: Fri, 19 Nov 2021 02:28:23 +0100
From: Ansuel Smith <ansuelsmth@...il.com>
To: Vladimir Oltean <olteanv@...il.com>
Cc: Andrew Lunn <andrew@...n.ch>,
Vivien Didelot <vivien.didelot@...il.com>,
Florian Fainelli <f.fainelli@...il.com>,
"David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>,
Russell King <linux@...linux.org.uk>,
linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [net-next PATCH 08/19] net: dsa: qca8k: convert qca8k to regmap
helper
On Fri, Nov 19, 2021 at 03:14:10AM +0200, Vladimir Oltean wrote:
> On Wed, Nov 17, 2021 at 10:04:40PM +0100, Ansuel Smith wrote:
> > Convert any qca8k read/write/rmw/set/clear/pool to regmap helper and add
> > missing config to regmap_config struct.
> >
> > Signed-off-by: Ansuel Smith <ansuelsmth@...il.com>
> > ---
>
> The important question is "why" and this commit message seems to omit that.
> Using regmap will be slower than using the equivalent direct I/O.
Yes sorry, will improve the message.
The transition to regmap is needed to permit the use of common code by
different switch that have different read/write/rmw function.
It seems cleaner to use regmap instead of using some helper or putting
the read/write/rmw in the priv struct.
Also in theory the overhead created by using regmap should be marginal
as the internal mdio use dedicated function and bypass regmap. So the
overhead should be present only in the configuration operation or fdb
access.
--
Ansuel
Powered by blists - more mailing lists