[PATCH] RTC: rtc-twl: Convert to module_platform_driver() and relocate reg_map init

[PATCH] RTC: rtc-twl: Convert to module_platform_driver() and relocate reg_map init

[Peter Ujfalusi]

Convert the driver to use module_platform_driver() to register the platform
driver and relocate the rtc_reg_map initialization to platform driver's
probe function.
In this way we can make sure that the twl-core has been already probed since
the core driver will create the device at the end of it's probe function.

Reported-by: Christoph Fritz <chf.fritz@googlemail.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
---
Hi Tony, Alessandro, Christoph,

I think this patch is the correct solution for the issue Christoph reported:
http://www.spinics.net/lists/linux-omap/msg89980.html

We can make sure that the twl-core has been already probed and initialized when
the RTC driver tries to pick the correct register map.

Regards,
Peter

 drivers/rtc/rtc-twl.c | 23 +++++++----------------
 1 file changed, 7 insertions(+), 16 deletions(-)

diff --git a/drivers/rtc/rtc-twl.c b/drivers/rtc/rtc-twl.c
index 8bc6c80..22b2fd6 100644
--- a/drivers/rtc/rtc-twl.c
+++ b/drivers/rtc/rtc-twl.c
@@ -469,6 +469,12 @@ static int twl_rtc_probe(struct platform_device *pdev)
 	if (irq <= 0)
 		goto out1;
 
+	/* Initialize the register map */
+	if (twl_class_is_4030())
+		rtc_reg_map = (u8 *)twl4030_rtc_reg_map;
+	else
+		rtc_reg_map = (u8 *)twl6030_rtc_reg_map;
+
 	ret = twl_rtc_read_u8(&rd_reg, REG_RTC_STATUS_REG);
 	if (ret < 0)
 		goto out1;
@@ -612,22 +618,7 @@ static struct platform_driver twl4030rtc_driver = {
 	},
 };
 
-static int __init twl_rtc_init(void)
-{
-	if (twl_class_is_4030())
-		rtc_reg_map = (u8 *) twl4030_rtc_reg_map;
-	else
-		rtc_reg_map = (u8 *) twl6030_rtc_reg_map;
-
-	return platform_driver_register(&twl4030rtc_driver);
-}
-module_init(twl_rtc_init);
-
-static void __exit twl_rtc_exit(void)
-{
-	platform_driver_unregister(&twl4030rtc_driver);
-}
-module_exit(twl_rtc_exit);
+module_platform_driver(twl4030rtc_driver);
 
 MODULE_AUTHOR("Texas Instruments, MontaVista Software");
 MODULE_LICENSE("GPL");
-- 
1.8.1.5

Re: [PATCH] RTC: rtc-twl: Convert to module_platform_driver() and relocate reg_map init

[Tomi Valkeinen]

Hi,

On 2013-04-16 10:44, Peter Ujfalusi wrote:
> Convert the driver to use module_platform_driver() to register the platform
> driver and relocate the rtc_reg_map initialization to platform driver's
> probe function.
> In this way we can make sure that the twl-core has been already probed since
> the core driver will create the device at the end of it's probe function.
> 
> Reported-by: Christoph Fritz <chf.fritz@googlemail.com>
> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>

I think there are two distinct things here. The actual fix for the
regmap, and a module_platform_driver cleanup. I would suggest having
them in separate patches.

And even if these are combined, I think the patch subject and
description should talk about fixing the regmap bug. Now it's rather
unclear that an actual bug is being fixed.

 Tomi

Re: [PATCH] RTC: rtc-twl: Convert to module_platform_driver() and relocate reg_map init

[Peter Ujfalusi]

Hi Tomi,

On 04/16/2013 10:10 AM, Tomi Valkeinen wrote:
> Hi,
> 
> On 2013-04-16 10:44, Peter Ujfalusi wrote:
>> Convert the driver to use module_platform_driver() to register the platform
>> driver and relocate the rtc_reg_map initialization to platform driver's
>> probe function.
>> In this way we can make sure that the twl-core has been already probed since
>> the core driver will create the device at the end of it's probe function.
>>
>> Reported-by: Christoph Fritz <chf.fritz@googlemail.com>
>> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
> 
> I think there are two distinct things here. The actual fix for the
> regmap, and a module_platform_driver cleanup. I would suggest having
> them in separate patches.
> 
> And even if these are combined, I think the patch subject and
> description should talk about fixing the regmap bug. Now it's rather
> unclear that an actual bug is being fixed.

Good point. My thinking was that when I convert the driver to
module_platform_driver() the reg map init must be moved to platform_driver's
probe anyways.
I'll separate the patch and write a bit better commit message for the first.

-- 
Péter