Date: Thu, 5 Feb 2015 23:40:40 +0000
From: Russell King - ARM Linux <linux@arm.linux.org.uk>
To: Wolfram Sang <wsa@the-dreams.de>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] clk: return proper ERR_PTR for clk_get when !HAVE_CLK
Message-ID: <20150205234040.GV8656@n2100.arm.linux.org.uk>
References: <1423163365-15267-1-git-send-email-wsa@the-dreams.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1423163365-15267-1-git-send-email-wsa@the-dreams.de>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1182
Lines: 29

On Thu, Feb 05, 2015 at 08:09:22PM +0100, Wolfram Sang wrote:
> clk_get functions return an ERR_PTR and not NULL in the error case. Make
> that consistent for the dummy functions when HAVE_CLK is not enabled.
> Otherwise unexpected codepaths might be trying to use a NULL pointer.

NAK.

There are some drivers which rely on this behaviour (take a driver,
such as some PXA IP) which uses the clk API but is also reused on
x86 which doesn't.

Remember, for any clk API implementation, it is well defined that
the clk API should safely eat any struct clk that clk_get() may
return which isn't a pointer-error.

So, if an implementation returns the NULL pointer, then the clk API
functions must accept it without crashing the kernel.

Also, clock consumer drivers are _not_ allowed to dereference
struct clk.  Absolutely not permitted.

-- 
FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up
according to speedtest.net.
--
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/