Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1674133 > unrolled thread

Re: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer

Started byAlexandre Belloni <alexandre.belloni@free-electrons.com>
First post2017-06-24 21:00 +0200
Last post2017-06-26 23:50 +0200
Articles 4 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer Alexandre Belloni <alexandre.belloni@free-electrons.com> - 2017-06-24 21:00 +0200
    Re: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer Florian Fainelli <f.fainelli@gmail.com> - 2017-06-26 19:20 +0200
      Re: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer Alexandre Belloni <alexandre.belloni@free-electrons.com> - 2017-06-26 23:40 +0200
        Re: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer Florian Fainelli <f.fainelli@gmail.com> - 2017-06-26 23:50 +0200

#1674133 — Re: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer

FromAlexandre Belloni <alexandre.belloni@free-electrons.com>
Date2017-06-24 21:00 +0200
SubjectRe: [PATCH 2/2] rtc: brcmstb-waketimer: Add Broadcom STB wake-timer
Message-ID<tVYrM-2Yp-31@gated-at.bofh.it>
Hi,

This seems mostly good.

On 15/06/2017 at 12:59:04 -0700, Florian Fainelli wrote:
> +static void wktmr_read(struct brcmstb_waketmr *timer,
> +		       struct wktmr_time *t)
> +{
> +	u32 tmp;
> +

To be sure, is this IP always 32bit, even on 64bit platforms?

> +	do {
> +		t->sec = readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER);
> +		tmp = readl_relaxed(timer->base + BRCMSTB_WKTMR_PRESCALER_VAL);
> +	} while (tmp >= timer->rate);
> +
> +	t->pre = timer->rate - tmp;
> +}
> +

[...]

> +static int brcmstb_waketmr_settime(struct device *dev,
> +				   struct rtc_time *tm)
> +{
> +	struct brcmstb_waketmr *timer = dev_get_drvdata(dev);
> +	unsigned long sec;
> +	int ret;
> +
> +	ret = rtc_valid_tm(tm);
> +	if (ret)
> +		return ret;
> +

There is no way this function can be called without a valid tm. The only
caller checks before calling.

> +	rtc_tm_to_time(tm, &sec);
> +
> +	dev_dbg(dev, "%s: sec=%ld\n", __func__, sec);
> +	writel_relaxed(sec, timer->base + BRCMSTB_WKTMR_COUNTER);
> +
> +	return 0;
> +}
> +
> +static int brcmstb_waketmr_getalarm(struct device *dev,
> +				    struct rtc_wkalrm *alarm)
> +{
> +	struct brcmstb_waketmr *timer = dev_get_drvdata(dev);
> +	unsigned long sec;
> +	u32 reg;
> +
> +	sec = readl_relaxed(timer->base + BRCMSTB_WKTMR_ALARM);
> +	if (sec == 0) {
> +		/* Alarm is disabled */
> +		alarm->enabled = 0;
> +		alarm->time.tm_mon = -1;
> +		alarm->time.tm_mday = -1;
> +		alarm->time.tm_year = -1;
> +		alarm->time.tm_hour = -1;
> +		alarm->time.tm_min = -1;
> +		alarm->time.tm_sec = -1;

This is not needed since d68778b80dd7

> +		dev_dbg(dev, "%s: alarm is disabled\n", __func__);
> +	} else {
> +		/* Alarm is enabled */
> +		alarm->enabled = 1;
> +		rtc_time_to_tm(sec, &alarm->time);
> +		dev_dbg(dev, "%s: alarm is enabled\n", __func__);
> +	}
> +
> +	reg = readl_relaxed(timer->base + BRCMSTB_WKTMR_EVENT);
> +	alarm->pending = !!(reg & 1);
> +	dev_dbg(dev, "%s: alarm pending=%d\n", __func__, alarm->pending);
> +
> +	return 0;
> +}

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

[toc] | [next] | [standalone]


#1674956

FromFlorian Fainelli <f.fainelli@gmail.com>
Date2017-06-26 19:20 +0200
Message-ID<tWFQ5-5au-13@gated-at.bofh.it>
In reply to#1674133
On 06/24/2017 11:59 AM, Alexandre Belloni wrote:
> Hi,
> 
> This seems mostly good.
> 
> On 15/06/2017 at 12:59:04 -0700, Florian Fainelli wrote:
>> +static void wktmr_read(struct brcmstb_waketmr *timer,
>> +		       struct wktmr_time *t)
>> +{
>> +	u32 tmp;
>> +
> 
> To be sure, is this IP always 32bit, even on 64bit platforms?

Correct, it's only 32-bit capable (saw the recent discussions about the
2038yr "problem"...).

> 
>> +	do {
>> +		t->sec = readl_relaxed(timer->base + BRCMSTB_WKTMR_COUNTER);
>> +		tmp = readl_relaxed(timer->base + BRCMSTB_WKTMR_PRESCALER_VAL);
>> +	} while (tmp >= timer->rate);
>> +
>> +	t->pre = timer->rate - tmp;
>> +}
>> +
> 
> [...]
> 
>> +static int brcmstb_waketmr_settime(struct device *dev,
>> +				   struct rtc_time *tm)
>> +{
>> +	struct brcmstb_waketmr *timer = dev_get_drvdata(dev);
>> +	unsigned long sec;
>> +	int ret;
>> +
>> +	ret = rtc_valid_tm(tm);
>> +	if (ret)
>> +		return ret;
>> +
> 
> There is no way this function can be called without a valid tm. The only
> caller checks before calling.

OK.

> 
>> +	rtc_tm_to_time(tm, &sec);
>> +
>> +	dev_dbg(dev, "%s: sec=%ld\n", __func__, sec);
>> +	writel_relaxed(sec, timer->base + BRCMSTB_WKTMR_COUNTER);
>> +
>> +	return 0;
>> +}
>> +
>> +static int brcmstb_waketmr_getalarm(struct device *dev,
>> +				    struct rtc_wkalrm *alarm)
>> +{
>> +	struct brcmstb_waketmr *timer = dev_get_drvdata(dev);
>> +	unsigned long sec;
>> +	u32 reg;
>> +
>> +	sec = readl_relaxed(timer->base + BRCMSTB_WKTMR_ALARM);
>> +	if (sec == 0) {
>> +		/* Alarm is disabled */
>> +		alarm->enabled = 0;
>> +		alarm->time.tm_mon = -1;
>> +		alarm->time.tm_mday = -1;
>> +		alarm->time.tm_year = -1;
>> +		alarm->time.tm_hour = -1;
>> +		alarm->time.tm_min = -1;
>> +		alarm->time.tm_sec = -1;
> 
> This is not needed since d68778b80dd7

Great, I will take that out.

Do you care whether some dev_dbg() prints are left in the driver or
should I remove those in v2?

> 
>> +		dev_dbg(dev, "%s: alarm is disabled\n", __func__);
>> +	} else {
>> +		/* Alarm is enabled */
>> +		alarm->enabled = 1;
>> +		rtc_time_to_tm(sec, &alarm->time);
>> +		dev_dbg(dev, "%s: alarm is enabled\n", __func__);
>> +	}
>> +
>> +	reg = readl_relaxed(timer->base + BRCMSTB_WKTMR_EVENT);
>> +	alarm->pending = !!(reg & 1);
>> +	dev_dbg(dev, "%s: alarm pending=%d\n", __func__, alarm->pending);
>> +
>> +	return 0;
>> +}
> 


-- 
Florian

[toc] | [prev] | [next] | [standalone]


#1675118

FromAlexandre Belloni <alexandre.belloni@free-electrons.com>
Date2017-06-26 23:40 +0200
Message-ID<tWJTI-7Ec-13@gated-at.bofh.it>
In reply to#1674956
On 26/06/2017 at 10:15:46 -0700, Florian Fainelli wrote:
> > This is not needed since d68778b80dd7
> 
> Great, I will take that out.
> 
> Do you care whether some dev_dbg() prints are left in the driver or
> should I remove those in v2?
> 

It could have stayed but it seems you removed it from v2.

I'll applay that but it is ok if you change your mind.

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

[toc] | [prev] | [next] | [standalone]


#1675124

FromFlorian Fainelli <f.fainelli@gmail.com>
Date2017-06-26 23:50 +0200
Message-ID<tWK3o-7Ki-33@gated-at.bofh.it>
In reply to#1675118
On 06/26/2017 02:35 PM, Alexandre Belloni wrote:
> On 26/06/2017 at 10:15:46 -0700, Florian Fainelli wrote:
>>> This is not needed since d68778b80dd7
>>
>> Great, I will take that out.
>>
>> Do you care whether some dev_dbg() prints are left in the driver or
>> should I remove those in v2?
>>
> 
> It could have stayed but it seems you removed it from v2.
> 
> I'll applay that but it is ok if you change your mind.

No, I think they don't really belong in the driver anymore, thanks!
-- 
Florian

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web