Date: Tue, 1 Mar 2011 08:59:11 +0000
From: Russell King - ARM Linux <linux@arm.linux.org.uk>
To: Simon Glass <sjg@chromium.org>
Cc: Nicolas Pitre <nicolas.pitre@linaro.org>,
        Phil Carmody <ext-phil.2.carmody@nokia.com>,
        Tony Lindgren <tony@atomide.com>,
        Catalin Marinas <catalin.marinas@arm.com>,
        linux-kernel@vger.kernel.org, Rabin Vincent <rabin@rab.in>,
        Alexander Shishkin <virtuoso@slind.org>,
        Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
        Joe Perches <joe@perches.com>, linux-arm-kernel@lists.infradead.org
Subject: Re: [RFC PATCH] ARM: Use generic BUG() handler
Message-ID: <20110301085911.GB16733@n2100.arm.linux.org.uk>
References: <1298939263-16421-1-git-send-email-sjg@chromium.org> <20110301084949.GA16733@n2100.arm.linux.org.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20110301084949.GA16733@n2100.arm.linux.org.uk>
User-Agent: Mutt/1.5.19 (2009-01-05)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1349
Lines: 36

On Tue, Mar 01, 2011 at 08:49:49AM +0000, Russell King - ARM Linux wrote:
> On Mon, Feb 28, 2011 at 04:27:43PM -0800, Simon Glass wrote:
> > +	asm volatile("1:\t.word %c3\n"				\
> > +		     ".pushsection __bug_table,\"a\"\n"		\
> > +		     "2:\t.word 1b, %c0\n"			\
> > +		     "\t.hword %c1, 0\n"			\
> > +		     "\t.org 2b+%c2\n"				\
> 
> %c doesn't work on lots of versions of gcc, which is why we can't use
> the generic bug support.  There's no way to reliably generate constants
> without many compiler versions spitting out a '#' before them.

gcc 4.3.2:

asm(".word %c0" : : "i" (0));

produces:

        .word #0

which gas chokes on:

/tmp/cc2hGOHd.s:12: Error: bad expression
/tmp/cc2hGOHd.s:12: Error: junk at end of line, first unrecognized character is `0'

So what this means is that it's impossible to generate constants in
assembly with GCC targetting ARM without having them prefixed by '#',
which in turn makes it impossible to use the generic BUG support.

I reported this bug to gcc folk many years ago.  I've no idea which
version it has been fixed in or if it's even been fixed.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/