MVME: Adopt standard RTC driver

Instead of hard-coded values and ifdefs, store the year offset in the platform_data struct.

Tested-by Daniel Palmer <[email protected]>

Reviewed-by Geert Uytterhoeven <[email protected]>

Signed-off-by Finn Thain <[email protected]>

---
David, Andreas - with your acknowledgement, I will ask Alexandre to merge this.
---
I tested this for regressions using qemu-system-sparc64. Also, Daniel tested the RTC functionality on his MVME147 system.

Changed since v2:
- Use an int for the year offset in struct m48t59_plat_data.

Changed since v3:
- Added reviewed-by tag from Geert.

---
 arch/sparc/kernel/time_32.c |  1 +
 arch/sparc/kernel/time_64.c |  1 +
 drivers/rtc/rtc-m48t59.c    | 26 ++++----------------------
 include/linux/rtc/m48t59.h  |  3 +++
 4 files changed, 9 insertions(+), 22 deletions(-)

arch/sparc/kernel/time_32.c +1 -0

index 08bbdc458596..578fd0d49f30 100644
@@ -255,6 +255,7 @@ static void mostek_write_byte(struct device *dev, u32 ofs, u8 val)
 static struct m48t59_plat_data m48t59_data = {
 	.read_byte = mostek_read_byte,
 	.write_byte = mostek_write_byte,
+	.yy_offset = 68,
 };
 
 /* resource is set at runtime */

arch/sparc/kernel/time_64.c +1 -0

index 60f1c8cc5363..b32f27f929d1 100644
@@ -544,6 +544,7 @@ static void mostek_write_byte(struct device *dev, u32 ofs, u8 val)
 static struct m48t59_plat_data m48t59_data = {
 	.read_byte	= mostek_read_byte,
 	.write_byte	= mostek_write_byte,
+	.yy_offset	= 68,
 };
 
 static struct platform_device m48t59_rtc = {

drivers/rtc/rtc-m48t59.c +4 -22

index 5d30ce8e13ca..4e608bc8bbd3 100644
@@ -71,7 +71,7 @@ static int m48t59_rtc_read_time(struct device *dev, struct rtc_time *tm)
 	/* Issue the READ command */
 	M48T59_SET_BITS(M48T59_CNTL_READ, M48T59_CNTL);
 
-	tm->tm_year	= bcd2bin(M48T59_READ(M48T59_YEAR));
+	tm->tm_year	= bcd2bin(M48T59_READ(M48T59_YEAR)) + pdata->yy_offset;
 	/* tm_mon is 0-11 */
 	tm->tm_mon	= bcd2bin(M48T59_READ(M48T59_MONTH)) - 1;
 	tm->tm_mday	= bcd2bin(M48T59_READ(M48T59_MDAY));

@@ -82,10 +82,6 @@ static int m48t59_rtc_read_time(struct device *dev, struct rtc_time *tm)
 		dev_dbg(dev, "Century bit is enabled\n");
 		tm->tm_year += 100;	/* one century */
 	}
-#ifdef CONFIG_SPARC
-	/* Sun SPARC machines count years since 1968 */
-	tm->tm_year += 68;
-#endif
 
 	tm->tm_wday	= bcd2bin(val & 0x07);
 	tm->tm_hour	= bcd2bin(M48T59_READ(M48T59_HOUR) & 0x3F);

@@ -106,12 +102,7 @@ static int m48t59_rtc_set_time(struct device *dev, struct rtc_time *tm)
 	struct m48t59_private *m48t59 = dev_get_drvdata(dev);
 	unsigned long flags;
 	u8 val = 0;
-	int year = tm->tm_year;
-
-#ifdef CONFIG_SPARC
-	/* Sun SPARC machines count years since 1968 */
-	year -= 68;
-#endif
+	int year = tm->tm_year - pdata->yy_offset;
 
 	dev_dbg(dev, "RTC set time %04d-%02d-%02d %02d/%02d/%02d\n",
 		year + 1900, tm->tm_mon, tm->tm_mday,

@@ -162,11 +153,7 @@ static int m48t59_rtc_readalarm(struct device *dev, struct rtc_wkalrm *alrm)
 	/* Issue the READ command */
 	M48T59_SET_BITS(M48T59_CNTL_READ, M48T59_CNTL);
 
-	tm->tm_year = bcd2bin(M48T59_READ(M48T59_YEAR));
-#ifdef CONFIG_SPARC
-	/* Sun SPARC machines count years since 1968 */
-	tm->tm_year += 68;
-#endif
+	tm->tm_year = bcd2bin(M48T59_READ(M48T59_YEAR)) + pdata->yy_offset;
 	/* tm_mon is 0-11 */
 	tm->tm_mon = bcd2bin(M48T59_READ(M48T59_MONTH)) - 1;
 

@@ -197,12 +184,7 @@ static int m48t59_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm)
 	struct rtc_time *tm = &alrm->time;
 	u8 mday, hour, min, sec;
 	unsigned long flags;
-	int year = tm->tm_year;
-
-#ifdef CONFIG_SPARC
-	/* Sun SPARC machines count years since 1968 */
-	year -= 68;
-#endif
+	int year = tm->tm_year - pdata->yy_offset;
 
 	/* If no irq, we don't support ALARM */
 	if (m48t59->irq == NO_IRQ)

include/linux/rtc/m48t59.h +0 -0

index 9465d5405fe2..373ba77071c6 100644

Instead of hard-coded values and ifdefs, store the year offset in the platform_data struct.

Quoted 4 lines · Finn Thain

Instead of hard-coded values and ifdefs, store the year offset in the platform_data struct.

Applied, thanks!

[1/2] rtc: m48t59: Use platform_data struct for year offset value

      https://git.kernel.org/abelloni/c/a06e4a93067c

Quoted 6 lines · Alexandre Belloni

Quoted 4 lines · Finn Thain

Instead of hard-coded values and ifdefs, store the year offset in the platform_data struct.

Quoted 5 lines

Applied, thanks!

[1/2] rtc: m48t59: Use platform_data struct for year offset value

      https://git.kernel.org/abelloni/c/a06e4a93067c

Thanks, Alexandre. Would you also take patch 2/2, please? Geert has sent a reviewed-by tag for that one too.

Quoted 7 lines · Finn Thain

Quoted 6 lines · Alexandre Belloni

Quoted 2 lines · Finn Thain

Instead of hard-coded values and ifdefs, store the year offset in the platform_data struct.

Applied, thanks!

Quoted 5 lines

Quoted 2 lines

[1/2] rtc: m48t59: Use platform_data struct for year offset value

      https://git.kernel.org/abelloni/c/a06e4a93067c

Thanks, Alexandre. Would you also take patch 2/2, please? Geert has sent a reviewed-by tag for that one too.

I thought Geert would take it as this only touches arch/m68k

Quoted 11 lines · Finn Thain

Quoted 8 lines · Alexandre Belloni

Quoted 4 lines · Finn Thain

Instead of hard-coded values and ifdefs, store the year offset in the platform_data struct.

Applied, thanks!

Quoted 6 lines

Quoted 3 lines

[1/2] rtc: m48t59: Use platform_data struct for year offset value

      https://git.kernel.org/abelloni/c/a06e4a93067c

Thanks, Alexandre. Would you also take patch 2/2, please? Geert has sent a reviewed-by tag for that one too.

I thought Geert would take it as this only touches arch/m68k

Both mvme147 and mvme16x platforms have their own RTC driver implementations that duplicate functionality provided by the rtc-m48t59 driver. Adopt the rtc-m48t59 driver and remove the other ones.

Tested-by Daniel Palmer <[email protected]>

Signed-off-by Finn Thain <[email protected]>

Reviewed-by Geert Uytterhoeven <[email protected]>

---
This patch depends upon the m48t59 driver changes in the preceding patch.

Changed since v1:
- Initialize yy_offset in struct m48t59_plat_data.

Changed Since v3:
- Re-ordered defconfig symbols.
- Added reviewed-by tag from arch maintainer.

Quoted 13 lines · Finn Thain

Both mvme147 and mvme16x platforms have their own RTC driver implementations that duplicate functionality provided by the rtc-m48t59 driver. Adopt the rtc-m48t59 driver and remove the other ones.

Tested-by Daniel Palmer <[email protected]>

Signed-off-by Finn Thain <[email protected]>

Reviewed-by Geert Uytterhoeven <[email protected]>

---
This patch depends upon the m48t59 driver changes in the preceding patch.

Changed since v1:
- Initialize yy_offset in struct m48t59_plat_data.

Quoted 3 lines

Changed Since v3:
- Re-ordered defconfig symbols.
- Added reviewed-by tag from arch maintainer.

Acked-by Geert Uytterhoeven <[email protected]>

so it still can make it into v6.13-rc1 via RTC?

Quoted 18 lines · Geert Uytterhoeven

Quoted 16 lines · Finn Thain

Both mvme147 and mvme16x platforms have their own RTC driver implementations that duplicate functionality provided by the rtc-m48t59 driver. Adopt the rtc-m48t59 driver and remove the other ones.

Tested-by Daniel Palmer <[email protected]>

Signed-off-by Finn Thain <[email protected]>

Reviewed-by Geert Uytterhoeven <[email protected]>

---
This patch depends upon the m48t59 driver changes in the preceding patch.

Changed since v1:
- Initialize yy_offset in struct m48t59_plat_data.

Changed Since v3:
- Re-ordered defconfig symbols.
- Added reviewed-by tag from arch maintainer.

Quoted 2 lines

Acked-by Geert Uytterhoeven <[email protected]>

so it still can make it into v6.13-rc1 via RTC?

It is now applied, it is going to conflict with your tree as it didn't cleanly apply. I don't think this is going to cause issues but I'll let it sit in linux-next for a few days before sending to Linus.