From mboxrd@z Thu Jan  1 00:00:00 1970
From: Nishanth Menon <menon.nishanth@gmail.com>
Date: Sun, 11 Apr 2010 08:09:34 -0500
Subject: [U-Boot] [PATCH 12/26] [ns16550] Enable port-mapped access
In-Reply-To: <4BC1C988.6040609@gmail.com>
References: <1270989829-2711-1-git-send-email-graeme.russ@gmail.com>
	<1270989829-2711-13-git-send-email-graeme.russ@gmail.com>
	<4BC1C5D5.3050200@gmail.com> <4BC1C988.6040609@gmail.com>
Message-ID: <4BC1CA0E.1010907@gmail.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

On 04/11/2010 08:07 AM, Graeme Russ wrote:
> Nishanth Menon wrote:
>> On 04/11/2010 07:43 AM, Graeme Russ wrote:
>>> The x86 architecture exclusively uses Port-Mapped I/O (inb/outb) to
>>> access
>>> the 16550 UARTs. This patch mimics how Linux selects between
>>> Memory-Mapped
>>> and Port-Mapped I/O. This allows x86 boards to use CONFIG_SERIAL_MUTLI
>>> and
>>> drop the custom serial port driver
>> [...]
>>
> <snip>
>>> +#ifdef CONFIG_X86
>>> +#define serial_out(x,y)    outb(x,(ulong)y)
>>> +#define serial_in(y)    inb((ulong)y)
>>> +#else
>>> +#define serial_out(x,y) writeb(x,y)
>>> +#define serial_in(y)     readb(y)
>>> +#endif
>>
>> wont it be nice if we have:
>> CONFIG_NS16550_PORT_MAPPED
>> CONFIG_NS16550_MEMORY_MAPPED
>> ? rather than being tied to x86?
> <snip>
>
> Hmm, true. Or maybe CONFIG_SYS_NS16550_PORT_MAPPED instead of
> CONFIG_X86 so the default of memory mapped is silently retained
>
> I think CONFIG_SYS is more appropriate as per the README:
>
> * Configuration _SETTINGS_:
>    These depend on the hardware etc. and should not be meddled with if
>    you don't know what you're doing; they have names beginning with
>    "CONFIG_SYS_".
>
That does sound good.. thanks..
Regards,
Nishanth Menon