From mboxrd@z Thu Jan  1 00:00:00 1970
From: u.kleine-koenig@pengutronix.de (Uwe =?iso-8859-1?Q?Kleine-K=F6nig?=)
Date: Tue, 17 Nov 2009 11:02:56 +0100
Subject: drivers/rtc/rtc-pl031.c uses __raw_readl()?
In-Reply-To: <63386a3d0911120043y2143c7f5w7289bc83b2e31847@mail.gmail.com>
References: <63386a3d0911110130j2056a355l16cade8d30edb6cd@mail.gmail.com>
	<63386a3d0911120043y2143c7f5w7289bc83b2e31847@mail.gmail.com>
Message-ID: <20091117100256.GC7407@pengutronix.de>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Thu, Nov 12, 2009 at 09:43:43AM +0100, Linus Walleij wrote:
> 2009/11/11 Linus Walleij <linus.ml.walleij@gmail.com>:
> 
> > Is there some special reason as to why drivers/rtc/rtc-pl031.c uses
> > the __raw_[read|write]l() macro instead of plain [read|write]l()?
> 
> Reading the macro definitions I come to the conclusion that readl()
> is always little-endian and __raw_readl() is the machine endianness,
Actually the readl part is not necessarily true I think.

At least arch/arm/mach-bcmring/include/mach/io.h has:

	/* Do not enable mem_pci for a big endian arm architecture or unexpected byteswaps will */
	/* happen in readw/writew etc. */

	#define readb(c)        __raw_readb(c)
	...

But maybe this is a bug in mach-bcmring?!  Or a work-around of drivers
using readb et al in wrong sitatations?

Best regards
Uwe

-- 
Pengutronix e.K.                              | Uwe Kleine-K?nig            |
Industrial Linux Solutions                    | http://www.pengutronix.de/  |