[rtc-linux] [PATCH 2/2] rtc: mediatek: Add MT63xx RTC driver

Eddie Huang eddie.huang at mediatek.com
Mon Mar 2 00:20:36 PST 2015


Hi Andrew,

On Mon, 2015-02-23 at 13:50 -0800, Andrew Morton wrote:
> On Wed, 28 Jan 2015 17:27:56 +0800 Eddie Huang <eddie.huang at mediatek.com> wrote:
> 
> > From: Tianping Fang <tianping.fang at mediatek.com>
> > 
> > Add Mediatek MT63xx RTC driver
> 
> There are a couple of checkpatch warnings which should be addressed,
> please:
> 
> WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
> #150: 
> new file mode 100644
> 
OK, I will update MAINTAINERS file

> WARNING: DT compatible string "mediatek,mt6397-rtc" appears un-documented -- check ./Documentation/devicetree/bindings/
> #488: FILE: drivers/rtc/rtc-mt6397.c:334:
> +	{ .compatible = "mediatek,mt6397-rtc", },
> 
> 
Do you include patch "[PATCH 1/2] dt: bindings: Add Mediatek RTC driver
binding document" in this series ?
http://lists.infradead.org/pipermail/linux-arm-kernel/2015-January/320425.html

I think this warning shouldn't happen if include this patch.


> >
> > ...
> >
> > +static void rtc_write(struct mt6397_rtc *rtc, u32 offset, u32 data)
> > +{
> > +	u32 addr;
> > +
> > +	addr = rtc->addr_base + offset;
> > +
> > +	if (offset < rtc->addr_range)
> > +		regmap_write(rtc->regmap, addr, data);
> > +}
> 
> regmap_read() and regmap_write() can return errors.  There is no
> checking for this.
Will fix it.

> > ...
> >
> > +static int mtk_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alm)
> > +{
> > +	struct rtc_time *tm = &alm->time;
> > +	struct mt6397_rtc *rtc = dev_get_drvdata(dev);
> > +	u16 irqen, pdn2;
> > +
> > +	mutex_lock(&rtc->lock);
> > +	irqen = rtc_read(rtc, RTC_IRQ_EN);
> > +	pdn2 = rtc_read(rtc, RTC_PDN2);
> > +	tm->tm_sec  = rtc_read(rtc, RTC_AL_SEC);
> > +	tm->tm_min  = rtc_read(rtc, RTC_AL_MIN);
> > +	tm->tm_hour = rtc_read(rtc, RTC_AL_HOU) & RTC_AL_HOU_MASK;
> > +	tm->tm_mday = rtc_read(rtc, RTC_AL_DOM) & RTC_AL_DOM_MASK;
> > +	tm->tm_mon  = rtc_read(rtc, RTC_AL_MTH) & RTC_AL_MTH_MASK;
> > +	tm->tm_year = rtc_read(rtc, RTC_AL_YEA);
> > +	mutex_unlock(&rtc->lock);
> > +
> > +	alm->enabled = !!(irqen & RTC_IRQ_EN_AL);
> > +	alm->pending = !!(pdn2 & RTC_PDN2_PWRON_ALARM);
> > +
> > +	tm->tm_year += RTC_MIN_YEAR_OFFSET;
> > +	tm->tm_mon--;
> > +
> > +	return 0;
> > +}
> > +
> > +static int mtk_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alm)
> > +{
> > +	struct rtc_time *tm = &alm->time;
> > +	struct mt6397_rtc *rtc = dev_get_drvdata(dev);
> > +	u16 irqen;
> > +
> > +	tm->tm_year -= RTC_MIN_YEAR_OFFSET;
> > +	tm->tm_mon++;
> > +
> > +	if (alm->enabled) {
> > +		mutex_lock(&rtc->lock);
> > +		rtc_write(rtc, RTC_AL_YEA, tm->tm_year);
> > +		rtc_write(rtc, RTC_AL_MTH, (rtc_read(rtc, RTC_AL_MTH) &
> > +				RTC_NEW_SPARE3) | tm->tm_mon);
> > +		rtc_write(rtc, RTC_AL_DOM, (rtc_read(rtc, RTC_AL_DOM) &
> > +				RTC_NEW_SPARE1) | tm->tm_mday);
> > +		rtc_write(rtc, RTC_AL_HOU, (rtc_read(rtc, RTC_AL_HOU) &
> > +				RTC_NEW_SPARE_FG_MASK) | tm->tm_hour);
> > +		rtc_write(rtc, RTC_AL_MIN, tm->tm_min);
> > +		rtc_write(rtc, RTC_AL_SEC, tm->tm_sec);
> > +		rtc_write(rtc, RTC_AL_MASK, RTC_AL_MASK_DOW);
> > +		rtc_write_trigger(rtc);
> > +		irqen = rtc_read(rtc, RTC_IRQ_EN) | RTC_IRQ_EN_ONESHOT_AL;
> > +		rtc_write(rtc, RTC_IRQ_EN, irqen);
> > +		rtc_write_trigger(rtc);
> > +		mutex_unlock(&rtc->lock);
> > +	}
> > +
> > +	return 0;
> > +}
> 
> This all looks a bit racy.  Wouldn't it be better if the testing of and
> modification of ->enabled and ->pending were protected by the mutex?
> 
Will fix it.

> >
> > ...
> >
> > +static int mtk_rtc_probe(struct platform_device *pdev)
> > +{
> > +	struct mt6397_chip *mt6397_chip = dev_get_drvdata(pdev->dev.parent);
> > +	struct mt6397_rtc *rtc;
> > +	u32 reg[2];
> > +	int ret = 0;
> > +
> > +	rtc = devm_kzalloc(&pdev->dev, sizeof(struct mt6397_rtc), GFP_KERNEL);
> > +	if (!rtc)
> > +		return -ENOMEM;
> > +
> > +	ret = of_property_read_u32_array(pdev->dev.of_node, "reg", reg, 2);
> > +	if (ret) {
> > +		dev_err(&pdev->dev, "couldn't read rtc base address!\n");
> > +		return -EINVAL;
> > +	}
> > +	rtc->addr_base = reg[0];
> > +	rtc->addr_range = reg[1];
> > +	rtc->regmap = mt6397_chip->regmap;
> > +	rtc->dev = &pdev->dev;
> > +	mutex_init(&rtc->lock);
> > +
> > +	platform_set_drvdata(pdev, rtc);
> > +
> > +	rtc->rtc_dev = rtc_device_register("mt6397-rtc", &pdev->dev,
> > +				&mtk_rtc_ops, THIS_MODULE);
> > +	if (IS_ERR(rtc->rtc_dev)) {
> > +		dev_err(&pdev->dev, "register rtc device failed\n");
> > +		return PTR_ERR(rtc->rtc_dev);
> > +	}
> > +
> > +	rtc->irq = platform_get_irq(pdev, 0);
> > +	if (rtc->irq < 0) {
> > +		ret = rtc->irq;
> > +		goto out_rtc;
> > +	}
> > +
> > +	ret = devm_request_threaded_irq(&pdev->dev, rtc->irq, NULL,
> > +			rtc_irq_handler_thread, IRQF_ONESHOT,
> > +			"mt6397-rtc", rtc);
> > +	if (ret) {
> > +		dev_err(&pdev->dev, "Failed to request alarm IRQ: %d: %d\n",
> > +			rtc->irq, ret);
> > +		goto out_rtc;
> > +	}
> > +
> > +	device_init_wakeup(&pdev->dev, 1);
> > +
> > +	return 0;
> > +
> > +out_rtc:
> > +	rtc_device_unregister(rtc->rtc_dev);
> > +	return ret;
> > +
> > +}
> 
> It seems strange to request the IRQ after having registered the rtc. 
> And possibly racy - I seem to recall another driver having issues with
> this recently.
> 
> A lot of rtc drivers are requesting the IRQ after registration so
> presumably it isn't a huge problem.  But still, wouldn't it be better
> to defer registration until after the IRQ has been successfully
> obtained?
> 
OK, will fix it.

Eddie





More information about the linux-arm-kernel mailing list