Patchwork rtc-omap: Fix initialization of control register

login
register
mail settings
Submitter Daniel Glöckner
Date July 11, 2011, 10:57 a.m.
Message ID <1310381866-25584-1-git-send-email-dg@emlix.com>
Download mbox | patch
Permalink /patch/104195/
State New
Headers show

Comments

Daniel Glöckner - July 11, 2011, 10:57 a.m.
As the comment explains, the intention of the code is to clear the
OMAP_RTC_CTRL_MODE_12_24 bit, but instead it only clears the
OMAP_RTC_CTRL_SPLIT and OMAP_RTC_CTRL_AUTO_COMP bits, which should be
kept. OMAP_RTC_CTRL_DISABLE, OMAP_RTC_CTRL_SET_32_COUNTER,
OMAP_RTC_CTRL_TEST, and OMAP_RTC_CTRL_ROUND_30S are also better off
being cleared.

Signed-off-by: Daniel Glöckner <dg@emlix.com>
---
 drivers/rtc/rtc-omap.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
Andrew Morton - July 27, 2011, 9:19 p.m.
On Mon, 11 Jul 2011 12:57:46 +0200
Daniel Gl__ckner <dg@emlix.com> wrote:

> As the comment explains, the intention of the code is to clear the
> OMAP_RTC_CTRL_MODE_12_24 bit, but instead it only clears the
> OMAP_RTC_CTRL_SPLIT and OMAP_RTC_CTRL_AUTO_COMP bits, which should be
> kept. OMAP_RTC_CTRL_DISABLE, OMAP_RTC_CTRL_SET_32_COUNTER,
> OMAP_RTC_CTRL_TEST, and OMAP_RTC_CTRL_ROUND_30S are also better off
> being cleared.
> 
> Signed-off-by: Daniel Gl__ckner <dg@emlix.com>
> ---
>  drivers/rtc/rtc-omap.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c
> index bcae8dd..7789002 100644
> --- a/drivers/rtc/rtc-omap.c
> +++ b/drivers/rtc/rtc-omap.c
> @@ -368,7 +368,7 @@ static int __init omap_rtc_probe(struct platform_device *pdev)
>  		pr_info("%s: already running\n", pdev->name);
>  
>  	/* force to 24 hour mode */
> -	new_ctrl = reg & ~(OMAP_RTC_CTRL_SPLIT|OMAP_RTC_CTRL_AUTO_COMP);
> +	new_ctrl = reg & (OMAP_RTC_CTRL_SPLIT|OMAP_RTC_CTRL_AUTO_COMP);
>  	new_ctrl |= OMAP_RTC_CTRL_STOP;
>  
>  	/* BOARD-SPECIFIC CUSTOMIZATION CAN GO HERE:

Decisions need to be made about which kernel version(s) this should be
merged into.  To make and support those decisions we need to know the
user-visible effects of the bug.  But the changelog is silent on this
important thing.

What were the user-visible effects of the bug?
Daniel Glöckner - July 28, 2011, 9:46 a.m.
On 07/27/2011 11:19 PM, Andrew Morton wrote:
> On Mon, 11 Jul 2011 12:57:46 +0200
> Daniel Gl__ckner <dg@emlix.com> wrote:
> 
>> As the comment explains, the intention of the code is to clear the
>> OMAP_RTC_CTRL_MODE_12_24 bit, but instead it only clears the
>> OMAP_RTC_CTRL_SPLIT and OMAP_RTC_CTRL_AUTO_COMP bits, which should be
>> kept. OMAP_RTC_CTRL_DISABLE, OMAP_RTC_CTRL_SET_32_COUNTER,
>> OMAP_RTC_CTRL_TEST, and OMAP_RTC_CTRL_ROUND_30S are also better off
>> being cleared.

> What were the user-visible effects of the bug?

The contents of the control register is undefined after a complete power
loss. On our board the RTC has been observed to sometimes power up with the
OMAP_RTC_CTRL_SET_32_COUNTER bit being set, which stops the clock.

  Daniel

Patch

diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c
index bcae8dd..7789002 100644
--- a/drivers/rtc/rtc-omap.c
+++ b/drivers/rtc/rtc-omap.c
@@ -368,7 +368,7 @@  static int __init omap_rtc_probe(struct platform_device *pdev)
 		pr_info("%s: already running\n", pdev->name);
 
 	/* force to 24 hour mode */
-	new_ctrl = reg & ~(OMAP_RTC_CTRL_SPLIT|OMAP_RTC_CTRL_AUTO_COMP);
+	new_ctrl = reg & (OMAP_RTC_CTRL_SPLIT|OMAP_RTC_CTRL_AUTO_COMP);
 	new_ctrl |= OMAP_RTC_CTRL_STOP;
 
 	/* BOARD-SPECIFIC CUSTOMIZATION CAN GO HERE: