Patchwork rtc: ds1307: Register offset moved to chip description

login
register
mail settings
Submitter Austin Boyle
Date June 26, 2012, 10:24 p.m.
Message ID <1340749442.3241.4046.camel@pc786-ubu.gnet.global.vpn>
Download mbox | patch
Permalink /patch/167469/
State New
Headers show

Comments

Austin Boyle - June 26, 2012, 10:24 p.m.
rtc: ds1307: Register offset moved to chip description

From: Austin Boyle <Austin.Boyle@aviatnet.com>

Register offset is added to chip description instead of being hard-coded
into probe function.

Based on v3.5-rc4

Cc: Wolfram Sang <w.sang@pengutronix.de>
Cc: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Austin Boyle <Austin.Boyle@aviatnet.com>

---
Wolfram Sang - June 27, 2012, 1:33 p.m.
On Wed, Jun 27, 2012 at 10:24:02AM +1200, Austin Boyle wrote:
> rtc: ds1307: Register offset moved to chip description
> 
> From: Austin Boyle <Austin.Boyle@aviatnet.com>
> 
> Register offset is added to chip description instead of being hard-coded
> into probe function.
> 
> Based on v3.5-rc4
> 
> Cc: Wolfram Sang <w.sang@pengutronix.de>
> Cc: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
> Cc: Alessandro Zummo <a.zummo@towertech.it>
> Signed-off-by: Austin Boyle <Austin.Boyle@aviatnet.com>

I would ack this patch immediately if it added another type using this
feature. But only changing already working code doesn't seem necessary
to me.

Regards,

    Wolfram
Joakim Tjernlund - June 27, 2012, 1:42 p.m.
> 
> On Wed, Jun 27, 2012 at 10:24:02AM +1200, Austin Boyle wrote:
> > rtc: ds1307: Register offset moved to chip description
> > 
> > From: Austin Boyle <Austin.Boyle@aviatnet.com>
> > 
> > Register offset is added to chip description instead of being hard-coded
> > into probe function.
> > 
> > Based on v3.5-rc4
> > 
> > Cc: Wolfram Sang <w.sang@pengutronix.de>
> > Cc: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
> > Cc: Alessandro Zummo <a.zummo@towertech.it>
> > Signed-off-by: Austin Boyle <Austin.Boyle@aviatnet.com>
> 
> I would ack this patch immediately if it added another type using this
> feature. But only changing already working code doesn't seem necessary
> to me.
> 
> Regards,
> 
>     Wolfram
> 

That was my gut feeling too.

  Jocke

Patch

diff --git a/a/drivers/rtc/rtc-ds1307.c b/b/drivers/rtc/rtc-ds1307.c
index 836710c..abf837b 100644
--- a/a/drivers/rtc/rtc-ds1307.c
+++ b/b/drivers/rtc/rtc-ds1307.c
@@ -122,6 +122,7 @@  struct ds1307 {
 
 struct chip_desc {
 	unsigned		alarm:1;
+	u8			offset;
 	u16			nvram_offset;
 	u16			nvram_size;
 	u16			trickle_charger_reg;
@@ -149,6 +150,9 @@  static const struct chip_desc chips[last_ds_type] = {
 	[ds_1388] = {
 		.trickle_charger_reg = 0x0a,
 	},
+	[ds_1388] = {
+		.offset		= 1, /* seconds field starts at 1 */
+	},
 	[ds_3231] = {
 		.alarm		= 1,
 	},
@@ -646,6 +650,7 @@  static int __devinit ds1307_probe(struct i2c_client *client,
 
 	ds1307->client	= client;
 	ds1307->type	= id->driver_data;
+	ds1307->offset	= chip->offset;
 
 	if (pdata && pdata->trickle_charger_setup && chip->trickle_charger_reg)
 		i2c_smbus_write_byte_data(client, chip->trickle_charger_reg,
@@ -768,9 +773,6 @@  static int __devinit ds1307_probe(struct i2c_client *client,
 						  hour);
 		}
 		break;
-	case ds_1388:
-		ds1307->offset = 1; /* Seconds starts at 1 */
-		break;
 	default:
 		break;
 	}