[PATCH] [WATCHDOG] mpc8xxx_wdt: fix modular build

[Anton Vorontsov]

This patch fixes following build error when mpc8xxx_wdt is selected to
build as a module:

drivers/watchdog/mpc8xxx_wdt.c:304: error: redefinition of '__inittest'
drivers/watchdog/mpc8xxx_wdt.c:298: error: previous definition of '__inittest' was here
drivers/watchdog/mpc8xxx_wdt.c:304: error: redefinition of 'init_module'
drivers/watchdog/mpc8xxx_wdt.c:298: error: previous definition of 'init_module' was here

Reported-by: Dave Jones <davej@xxxxxxxxxx>
Signed-off-by: Anton Vorontsov <avorontsov@xxxxxxxxxxxxx>
---

On Tue, Aug 12, 2008 at 11:10:44PM -0400, Dave Jones wrote:
> On Wed, Aug 06, 2008 at 10:29:18PM +0200, Wim Van Sebroeck wrote:
>  > Hi Linus,
>  > 
>  > Please pull from 'master' branch of
>  > 	git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog.git
>  > or if master.kernel.org hasn't synced up yet:
>  > 	master.kernel.org:/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog.git
>  > 
>  > This will update the following files:
>  > 
>  > Author: Anton Vorontsov <avorontsov@xxxxxxxxxxxxx>
>  > Date:   Thu Jul 3 23:51:36 2008 -0700
>  > 
>  >     [WATCHDOG] mpc8xxx_wdt: add support for MPC8xx watchdogs
> 
> This is broken when built modular because this...
> 
>  > +}
>  > +module_init(mpc8xxx_wdt_init_late);
> 
> and this ..
> 
>  > +arch_initcall(mpc8xxx_wdt_init);
> 
> Can't be mixed.  You end up with two init_module invocations
> 
> drivers/watchdog/mpc8xxx_wdt.c:304: error: redefinition of '__inittest'
> drivers/watchdog/mpc8xxx_wdt.c:298: error: previous definition of '__inittest' was here
> drivers/watchdog/mpc8xxx_wdt.c:304: error: redefinition of 'init_module'
> drivers/watchdog/mpc8xxx_wdt.c:298: error: previous definition of 'init_module' was here
> 
> Given they both call different routines though, I'm not sure the right fix here.

Thanks for noticing!

Unfortunately I don't see any elegant way to fix this. In the long term,
it would be great if we could work with misc (thus watchdog) subsystem at
arch_initcall() time (yes, we need this early on 8xx). Until this happen,
I think this patch is the best fix we can do.

 drivers/watchdog/mpc8xxx_wdt.c |   11 ++++++++++-
 1 files changed, 10 insertions(+), 1 deletions(-)

diff --git a/drivers/watchdog/mpc8xxx_wdt.c b/drivers/watchdog/mpc8xxx_wdt.c
index f209496..8e50bc4 100644
--- a/drivers/watchdog/mpc8xxx_wdt.c
+++ b/drivers/watchdog/mpc8xxx_wdt.c
@@ -48,6 +48,7 @@ struct mpc8xxx_wdt_type {
 };
 
 static struct mpc8xxx_wdt __iomem *wd_base;
+static int mpc8xxx_wdt_init_late(void);
 
 static u16 timeout = 0xffff;
 module_param(timeout, ushort, 0);
@@ -213,6 +214,12 @@ static int __devinit mpc8xxx_wdt_probe(struct of_device *ofdev,
 	else
 		timeout_sec = timeout / freq;
 
+#ifdef CONFIG_8xxx_WDT_MODULE
+	ret = mpc8xxx_wdt_init_late();
+	if (ret)
+		goto err_unmap;
+#endif
+
 	pr_info("WDT driver for MPC8xxx initialized. mode:%s timeout=%d "
 		"(%d seconds)\n", reset ? "reset" : "interrupt", timeout,
 		timeout_sec);
@@ -280,7 +287,7 @@ static struct of_platform_driver mpc8xxx_wdt_driver = {
  * very early to start pinging the watchdog (misc devices are not yet
  * available), and later module_init() just registers the misc device.
  */
-static int __init mpc8xxx_wdt_init_late(void)
+static int mpc8xxx_wdt_init_late(void)
 {
 	int ret;
 
@@ -295,7 +302,9 @@ static int __init mpc8xxx_wdt_init_late(void)
 	}
 	return 0;
 }
+#ifndef CONFIG_8xxx_WDT_MODULE
 module_init(mpc8xxx_wdt_init_late);
+#endif
 
 static int __init mpc8xxx_wdt_init(void)
 {
-- 
1.5.6.3

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/