[PATCH 1/5] drivers/rtc/pcf8563: Replace deprecated rtc_time_to_tm() and rtc_tm_to_time()

[PATCH 1/5] drivers/rtc/pcf8563: Replace deprecated rtc_time_to_tm() and rtc_tm_to_time()

[Xunlei Pang]

From: Xunlei Pang <pang.xunlei@linaro.org>

pcf8563_rtc_set_alarm() uses deprecated rtc_tm_to_time()
and rtc_time_to_tm(), which will overflow in year 2106
on 32-bit machines.

This patch solves this by:
 - Replacing rtc_time_to_tm() with rtc_time64_to_tm()
 - Replacing rtc_tm_to_time() with rtc_tm_to_time64()

Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
---
 drivers/rtc/rtc-pcf8563.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/rtc/rtc-pcf8563.c b/drivers/rtc/rtc-pcf8563.c
index 96fb32e..4ce66ad 100644
--- a/drivers/rtc/rtc-pcf8563.c
+++ b/drivers/rtc/rtc-pcf8563.c
@@ -368,13 +368,13 @@ static int pcf8563_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *tm)
 	struct i2c_client *client = to_i2c_client(dev);
 	unsigned char buf[4];
 	int err;
-	unsigned long alarm_time;
 
 	/* The alarm has no seconds, round up to nearest minute */
 	if (tm->time.tm_sec) {
-		rtc_tm_to_time(&tm->time, &alarm_time);
-		alarm_time += 60-tm->time.tm_sec;
-		rtc_time_to_tm(alarm_time, &tm->time);
+		time64_t alarm_time = rtc_tm_to_time64(&tm->time);
+
+		alarm_time += 60 - tm->time.tm_sec;
+		rtc_time64_to_tm(alarm_time, &tm->time);
 	}
 
 	dev_dbg(dev, "%s, min=%d hour=%d wday=%d mday=%d "
-- 
1.9.1

[PATCH 2/5] drivers/rtc/isl1208: Replace deprecated rtc_tm_to_time()

[Xunlei Pang]

From: Xunlei Pang <pang.xunlei@linaro.org>

isl1208_i2c_set_alarm() uses deprecated rtc_tm_to_time(),
which will overflow in year 2106 on 32-bit machines.

This patch solves this by:
 - Replacing rtc_tm_to_time() with rtc_tm_to_time64()

Cc: Herbert Valerio Riedel <hvr@gnu.org>
Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
---
 drivers/rtc/rtc-isl1208.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/rtc/rtc-isl1208.c b/drivers/rtc/rtc-isl1208.c
index c3c549d..06113e8 100644
--- a/drivers/rtc/rtc-isl1208.c
+++ b/drivers/rtc/rtc-isl1208.c
@@ -370,19 +370,16 @@ isl1208_i2c_set_alarm(struct i2c_client *client, struct rtc_wkalrm *alarm)
 	struct rtc_time *alarm_tm = &alarm->time;
 	u8 regs[ISL1208_ALARM_SECTION_LEN] = { 0, };
 	const int offs = ISL1208_REG_SCA;
-	unsigned long rtc_secs, alarm_secs;
+	time64_t rtc_secs, alarm_secs;
 	struct rtc_time rtc_tm;
 	int err, enable;
 
 	err = isl1208_i2c_read_time(client, &rtc_tm);
 	if (err)
 		return err;
-	err = rtc_tm_to_time(&rtc_tm, &rtc_secs);
-	if (err)
-		return err;
-	err = rtc_tm_to_time(alarm_tm, &alarm_secs);
-	if (err)
-		return err;
+
+	rtc_secs = rtc_tm_to_time64(&rtc_tm);
+	alarm_secs = rtc_tm_to_time64(alarm_tm);
 
 	/* If the alarm time is before the current time disable the alarm */
 	if (!alarm->enabled || alarm_secs <= rtc_secs)
-- 
1.9.1

[PATCH 3/5] drivers/rtc/sunxi: Replace deprecated rtc_tm_to_time()

[Xunlei Pang]

From: Xunlei Pang <pang.xunlei@linaro.org>

sunxi_rtc_setalarm() uses deprecated rtc_tm_to_time(),
which will overflow in year 2106 on 32-bit machines.

This patch solves this by:
- Replacing rtc_tm_to_time() with rtc_tm_to_time64()

Also remove the unnecessary initial zeroing of some
local variables in sunxi_rtc_setalarm().

Cc: Carlo Caione <carlo.caione@gmail.com>
Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
---
 drivers/rtc/rtc-sunxi.c | 29 ++++++++++++++---------------
 1 file changed, 14 insertions(+), 15 deletions(-)

diff --git a/drivers/rtc/rtc-sunxi.c b/drivers/rtc/rtc-sunxi.c
index 6e678fa..7f22753 100644
--- a/drivers/rtc/rtc-sunxi.c
+++ b/drivers/rtc/rtc-sunxi.c
@@ -269,14 +269,13 @@ static int sunxi_rtc_setalarm(struct device *dev, struct rtc_wkalrm *wkalrm)
 	struct sunxi_rtc_dev *chip = dev_get_drvdata(dev);
 	struct rtc_time *alrm_tm = &wkalrm->time;
 	struct rtc_time tm_now;
-	u32 alrm = 0;
-	unsigned long time_now = 0;
-	unsigned long time_set = 0;
-	unsigned long time_gap = 0;
-	unsigned long time_gap_day = 0;
-	unsigned long time_gap_hour = 0;
-	unsigned long time_gap_min = 0;
-	int ret = 0;
+	u32 alrm;
+	time64_t time_set, time_now;
+	unsigned long time_gap;
+	unsigned long time_gap_day;
+	unsigned long time_gap_hour;
+	unsigned long time_gap_min;
+	int ret;
 
 	ret = sunxi_rtc_gettime(dev, &tm_now);
 	if (ret < 0) {
@@ -284,13 +283,18 @@ static int sunxi_rtc_setalarm(struct device *dev, struct rtc_wkalrm *wkalrm)
 		return -EINVAL;
 	}
 
-	rtc_tm_to_time(alrm_tm, &time_set);
-	rtc_tm_to_time(&tm_now, &time_now);
+	time_set = rtc_tm_to_time64(alrm_tm);
+	time_now = rtc_tm_to_time64(&tm_now);
 	if (time_set <= time_now) {
 		dev_err(dev, "Date to set in the past\n");
 		return -EINVAL;
 	}
 
+	if (time_set > time_now + 255 * SEC_IN_DAY) {
+		dev_err(dev, "Day must be in the range 0 - 255\n");
+		return -EINVAL;
+	}
+
 	time_gap = time_set - time_now;
 	time_gap_day = time_gap / SEC_IN_DAY;
 	time_gap -= time_gap_day * SEC_IN_DAY;
@@ -299,11 +303,6 @@ static int sunxi_rtc_setalarm(struct device *dev, struct rtc_wkalrm *wkalrm)
 	time_gap_min = time_gap / SEC_IN_MIN;
 	time_gap -= time_gap_min * SEC_IN_MIN;
 
-	if (time_gap_day > 255) {
-		dev_err(dev, "Day must be in the range 0 - 255\n");
-		return -EINVAL;
-	}
-
 	sunxi_rtc_setaie(0, chip);
 	writel(0, chip->base + SUNXI_ALRM_DHMS);
 	usleep_range(100, 300);
-- 
1.9.1

[PATCH 4/5] drivers/rtc/pl030: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Xunlei Pang]

From: Xunlei Pang <pang.xunlei@linaro.org>

The driver uses deprecated rtc_tm_to_time() and rtc_time_to_tm(),
which will overflow in year 2106 on 32-bit machines.

This patch solves this by:
 - Replacing rtc_tm_to_time() with rtc_tm_to_time64()
 - Replacing rtc_time_to_tm() with rtc_time64_to_tm()

Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
---
 drivers/rtc/rtc-pl030.c | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/drivers/rtc/rtc-pl030.c b/drivers/rtc/rtc-pl030.c
index f85a1a9..d874b6e 100644
--- a/drivers/rtc/rtc-pl030.c
+++ b/drivers/rtc/rtc-pl030.c
@@ -39,32 +39,34 @@ static int pl030_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 {
 	struct pl030_rtc *rtc = dev_get_drvdata(dev);
 
-	rtc_time_to_tm(readl(rtc->base + RTC_MR), &alrm->time);
+	rtc_time64_to_tm(readl(rtc->base + RTC_MR), &alrm->time);
 	return 0;
 }
 
 static int pl030_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 {
 	struct pl030_rtc *rtc = dev_get_drvdata(dev);
-	unsigned long time;
+	time64_t time;
 	int ret;
 
 	/*
 	 * At the moment, we can only deal with non-wildcarded alarm times.
 	 */
 	ret = rtc_valid_tm(&alrm->time);
-	if (ret == 0)
-		ret = rtc_tm_to_time(&alrm->time, &time);
-	if (ret == 0)
-		writel(time, rtc->base + RTC_MR);
-	return ret;
+	if (ret)
+		return ret;
+
+	time = rtc_tm_to_time64(&alrm->time);
+	writel(time, rtc->base + RTC_MR);
+
+	return 0;
 }
 
 static int pl030_read_time(struct device *dev, struct rtc_time *tm)
 {
 	struct pl030_rtc *rtc = dev_get_drvdata(dev);
 
-	rtc_time_to_tm(readl(rtc->base + RTC_DR), tm);
+	rtc_time64_to_tm(readl(rtc->base + RTC_DR), tm);
 
 	return 0;
 }
@@ -80,14 +82,12 @@ static int pl030_read_time(struct device *dev, struct rtc_time *tm)
 static int pl030_set_time(struct device *dev, struct rtc_time *tm)
 {
 	struct pl030_rtc *rtc = dev_get_drvdata(dev);
-	unsigned long time;
-	int ret;
+	time64_t time;
 
-	ret = rtc_tm_to_time(tm, &time);
-	if (ret == 0)
-		writel(time + 1, rtc->base + RTC_LR);
+	time = rtc_tm_to_time64(tm);
+	writel(time + 1, rtc->base + RTC_LR);
 
-	return ret;
+	return 0;
 }
 
 static const struct rtc_class_ops pl030_ops = {
-- 
1.9.1

Re: [PATCH 4/5] drivers/rtc/pl030: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Russell King - ARM Linux]

On Wed, Apr 15, 2015 at 05:20:10PM +0800, Xunlei Pang wrote:
> From: Xunlei Pang <pang.xunlei@linaro.org>
> 
> The driver uses deprecated rtc_tm_to_time() and rtc_time_to_tm(),
> which will overflow in year 2106 on 32-bit machines.
> 
> This patch solves this by:
>  - Replacing rtc_tm_to_time() with rtc_tm_to_time64()
>  - Replacing rtc_time_to_tm() with rtc_time64_to_tm()
> 
> Cc: Russell King <rmk+kernel@arm.linux.org.uk>
> Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>

NAK.

How does this fix anything?  The RTC contains 32-bit registers.  Even
if you convert the struct tm to a 64-bit time, you can only write the
lowest 32-bits to the hardware.  You can onyl read the lowest 32-bits
from the hardware too.

This patch solves /nothing/.  In fact, it hides the fact that this RTC
is unable to represent dates after 2106.

-- 
FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up
according to speedtest.net.

[PATCH 5/5] drivers/rtc/sa1100: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Xunlei Pang]

From: Xunlei Pang <pang.xunlei@linaro.org>

The driver uses deprecated rtc_tm_to_time() and rtc_time_to_tm(),
which will overflow in year 2106 on 32-bit machines.

This patch solves this by:
 - Replacing rtc_tm_to_time() with rtc_tm_to_time64()
 - Replacing rtc_time_to_tm() with rtc_time64_to_tm()

Cc: CIH <cih@coventive.com>
Cc: Nicolas Pitre <nico@fluxnic.net>
Cc: Andrew Christian <andrew.christian@hp.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
---
 drivers/rtc/rtc-sa1100.c | 21 +++++----------------
 1 file changed, 5 insertions(+), 16 deletions(-)

diff --git a/drivers/rtc/rtc-sa1100.c b/drivers/rtc/rtc-sa1100.c
index b6e1ca0..625a320 100644
--- a/drivers/rtc/rtc-sa1100.c
+++ b/drivers/rtc/rtc-sa1100.c
@@ -157,19 +157,14 @@ static int sa1100_rtc_alarm_irq_enable(struct device *dev, unsigned int enabled)
 
 static int sa1100_rtc_read_time(struct device *dev, struct rtc_time *tm)
 {
-	rtc_time_to_tm(RCNR, tm);
+	rtc_time64_to_tm(RCNR, tm);
 	return 0;
 }
 
 static int sa1100_rtc_set_time(struct device *dev, struct rtc_time *tm)
 {
-	unsigned long time;
-	int ret;
-
-	ret = rtc_tm_to_time(tm, &time);
-	if (ret == 0)
-		RCNR = time;
-	return ret;
+	RCNR = rtc_tm_to_time64(tm);
+	return 0;
 }
 
 static int sa1100_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
@@ -185,23 +180,17 @@ static int sa1100_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 static int sa1100_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
 {
 	struct sa1100_rtc *info = dev_get_drvdata(dev);
-	unsigned long time;
-	int ret;
 
 	spin_lock_irq(&info->lock);
-	ret = rtc_tm_to_time(&alrm->time, &time);
-	if (ret != 0)
-		goto out;
 	RTSR = RTSR & (RTSR_HZE|RTSR_ALE|RTSR_AL);
-	RTAR = time;
+	RTAR = rtc_tm_to_time64(&alrm->time);
 	if (alrm->enabled)
 		RTSR |= RTSR_ALE;
 	else
 		RTSR &= ~RTSR_ALE;
-out:
 	spin_unlock_irq(&info->lock);
 
-	return ret;
+	return 0;
 }
 
 static int sa1100_rtc_proc(struct device *dev, struct seq_file *seq)
-- 
1.9.1

Re: [5/5] drivers/rtc/sa1100: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Alexandre Belloni]

Hi,

On 15/04/2015 at 17:20:11 +0800, Xunlei Pang wrote :
> From: Xunlei Pang <pang.xunlei@linaro.org>
> 
> The driver uses deprecated rtc_tm_to_time() and rtc_time_to_tm(),
> which will overflow in year 2106 on 32-bit machines.
> 
> This patch solves this by:
>  - Replacing rtc_tm_to_time() with rtc_tm_to_time64()
>  - Replacing rtc_time_to_tm() with rtc_time64_to_tm()
> 
> Cc: CIH <cih@coventive.com>
> Cc: Nicolas Pitre <nico@fluxnic.net>
> Cc: Andrew Christian <andrew.christian@hp.com>
> Cc: Richard Purdie <rpurdie@rpsys.net>
> Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
> ---
>  drivers/rtc/rtc-sa1100.c | 21 +++++----------------
>  1 file changed, 5 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/rtc/rtc-sa1100.c b/drivers/rtc/rtc-sa1100.c
> index b6e1ca0..625a320 100644
> --- a/drivers/rtc/rtc-sa1100.c
> +++ b/drivers/rtc/rtc-sa1100.c
> @@ -157,19 +157,14 @@ static int sa1100_rtc_alarm_irq_enable(struct device *dev, unsigned int enabled)
>  
>  static int sa1100_rtc_read_time(struct device *dev, struct rtc_time *tm)
>  {
> -	rtc_time_to_tm(RCNR, tm);
> +	rtc_time64_to_tm(RCNR, tm);
>  	return 0;
>  }
>  
>  static int sa1100_rtc_set_time(struct device *dev, struct rtc_time *tm)
>  {
> -	unsigned long time;
> -	int ret;
> -
> -	ret = rtc_tm_to_time(tm, &time);
> -	if (ret == 0)
> -		RCNR = time;
> -	return ret;
> +	RCNR = rtc_tm_to_time64(tm);

For the same reason that Russell pointed in patch 4/5, this hides that
it doesn't work after 2106-02-07 06:28:16 as the register is still 32
bits.
I would prefer that you return an error in that case.


-- 
Alexandre Belloni, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Re: [5/5] drivers/rtc/sa1100: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Arnd Bergmann]

On Saturday 30 May 2015 18:05:10 pang.xunlei@zte.com.cn wrote:
> PangXunLei10110855/user/zte_ltd wrote 2015-05-29 PM 10:32:15:
> > Alexandre Belloni <alexandre.belloni@free-electrons.com> wrote 
> > 2015-04-30 AM 07:28:24:
> > > Re: [5/5] drivers/rtc/sa1100: Replace deprecated rtc_tm_to_time() 
> > > and rtc_time_to_tm()
> > > 
> > > For the same reason that Russell pointed in patch 4/5, this hides that
> > > it doesn't work after 2106-02-07 06:28:16 as the register is still 32
> > > bits.
> > > I would prefer that you return an error in that case.
> > 
> > Thanks for the review, I want to leave the ones like this untouched.
> > 
> 
> To further explain, for the rtc hardware with 32-bit counters, we actually 
> 
> had a solution before: https://lkml.org/lkml/2014/11/27/341
> 
> But seems some guys don't like the solution, so rtc drivers like this one 
> can't be simply changed to use the new-added y2038-safe interfaces.
> 
> Maybe those drivers will still retain the deprecated interfaces until a 
> better solution is proposed.

Note that Alexandre has stepped up as a new maintainer for RTC now,
so whatever he wants is probably the way it should be done.

I think the way that your first approach came out was a bit unfortunate,
because we didn't have agreement on how it should really be done.

IMHO the rtc_time64_to_hw32 and rtc_hw32_to_time64 interfaces were a
good concept, most importantly so we can easily find where the potential
problems are, but the implementation was a bit too ambitious in trying
to fix the underlying issue.

If we want to start this over again, I think a better approach would be
to introduce trivial functions at first, like

time64_t rtc_hw32_to_time64(u32 hwtime)
{
	/* 
	 * this is safe until about 2106, when unsigned u32 seconds from the
	 * 1970 epoch will overflow
	 */
	return (u64)hwtime;
}
EXPORT_SYMBOL_GPL(rtc_hw32_to_time64);

This way, we can fix all the drivers without introducing any possible
ambiguity and later decide how that function should in fact handle the
2106 overflow. Our grandchildren can take care of that if necessary ;-)

	Arnd

Re: [5/5] drivers/rtc/sa1100: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Alexandre Belloni]

Hi,

On 01/06/2015 at 21:43:22 +0200, Arnd Bergmann wrote :
> Note that Alexandre has stepped up as a new maintainer for RTC now,
> so whatever he wants is probably the way it should be done.
> 

I'm still quite new in that position and I didn't make my mind in a lot
of different topics ;)

> I think the way that your first approach came out was a bit unfortunate,
> because we didn't have agreement on how it should really be done.
> 
> IMHO the rtc_time64_to_hw32 and rtc_hw32_to_time64 interfaces were a
> good concept, most importantly so we can easily find where the potential
> problems are, but the implementation was a bit too ambitious in trying
> to fix the underlying issue.
> 
> If we want to start this over again, I think a better approach would be
> to introduce trivial functions at first, like
> 
> time64_t rtc_hw32_to_time64(u32 hwtime)
> {
> 	/* 
> 	 * this is safe until about 2106, when unsigned u32 seconds from the
> 	 * 1970 epoch will overflow
> 	 */
> 	return (u64)hwtime;
> }
> EXPORT_SYMBOL_GPL(rtc_hw32_to_time64);
> 
> This way, we can fix all the drivers without introducing any possible
> ambiguity and later decide how that function should in fact handle the
> 2106 overflow. Our grandchildren can take care of that if necessary ;-)
> 

We also have RTCs that can only store the year as an integer between 0
and 99. Some of them will definitively not pass 2099, considering that
the can only handle leap days between 2000 and 2099. Also, we have a few
drivers assuming that year >= 70 is in the 19s so they will stop working
correctly in 2070. I think we also have to consider those while trying
to find a solution.

I feel that we still have a bit of time before having to hurry and find
a proper solution ;)

-- 
Alexandre Belloni, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Re: [5/5] drivers/rtc/sa1100: Replace deprecated rtc_tm_to_time() and rtc_time_to_tm()

[Alessandro Zummo]

On Mon, 1 Jun 2015 21:59:09 +0200
Alexandre Belloni <alexandre.belloni@free-electrons.com> wrote:

> I feel that we still have a bit of time before having to hurry and find
> a proper solution ;)

 agreed :)

-- 

 Best regards,

 Alessandro Zummo - CEO,
  Tower Technologies - Torino, Italy

  http://www.towertech.it