Re: [PATCH] RM9000 serial driver

From: Thomas Koeller <thomas.koeller@baslerweb.com>
To: "Russell King" <rmk@arm.linux.org.uk>
Cc: "Sergei Shtylyov" <sshtylyov@ru.mvista.com>,
	"Yoichi Yuasa" <yoichi_yuasa@tripeaks.co.jp>,
	linux-serial@vger.kernel.org, ralf@linux-mips.org,
	linux-mips@linux-mips.org,
	"Thomas Köller" <thomas@koeller.dyndns.org>
Subject: Re: [PATCH] RM9000 serial driver
Date: Wed, 30 Aug 2006 23:16:42 +0200	[thread overview]
Message-ID: <200608302316.43111.thomas.koeller@baslerweb.com> (raw)
In-Reply-To: <20060829190426.GA20606@flint.arm.linux.org.uk>

On Tuesday 29 August 2006 21:04, Russell King wrote:
> It's worse than that - this code is there to read the ID from the divisor
> registers implemented in some UARTs.  If it isn't one of those UARTs, it's
> expected to return zero.
>
> So we don't actually want to be prodding some other random registers on
> differing UARTs.

For the  RM9000, DLL and DLM are located at distinct addresses, so these 
registers could be accessed without prior setting of DLAB. However, the
h/w docs say that DLAB has to be used nonetheless. I doubt that, but
wanted to play safe. So, in order to make this work, I had two options:

1. to monitor all register writes for setting/clearing of DLAB, and 
   switch the register mapping tables accordingly, or

2. implement serial_dl_read()/serial_dl_write() to directly access
   the registers, thus bypassing the mapping.

I decided to implement option #2, because it seemed less of a kludge.
Would you still say that this is an abuse?

Thomas
-- 
Thomas Koeller, Software Development

Basler Vision Technologies
An der Strusbek 60-62
22926 Ahrensburg
Germany

Tel +49 (4102) 463-390
Fax +49 (4102) 463-46390

mailto:thomas.koeller@baslerweb.com
http://www.baslerweb.com