Re: [PATCH v2 01/10] mfd: max77686: Convert to use regmap_irq

From: Javier Martinez Canillas
Date: Wed Jun 18 2014 - 05:22:16 EST


Hello Doug,

On 06/17/2014 10:57 PM, Doug Anderson wrote:
> Javier,
>
> On Mon, Jun 16, 2014 at 11:02 AM, Javier Martinez Canillas
> <javier.martinez@xxxxxxxxxxxxxxx> wrote:
>> @@ -127,15 +175,48 @@ static int max77686_i2c_probe(struct i2c_client *i2c,
>> }
>> i2c_set_clientdata(max77686->rtc, max77686);
>>
>> - max77686_irq_init(max77686);
>> + max77686->rtc_regmap = devm_regmap_init_i2c(max77686->rtc,
>> + &max77686_rtc_regmap_config);
>> + if (IS_ERR(max77686->rtc_regmap)) {
>> + ret = PTR_ERR(max77686->rtc_regmap);
>> + dev_err(max77686->dev, "failed to allocate RTC regmap: %d\n",
>> + ret);
>> + goto err_unregister_i2c;
>> + }
>> +
>> + ret = regmap_add_irq_chip(max77686->regmap, max77686->irq,
>> + IRQF_TRIGGER_FALLING | IRQF_ONESHOT |
>> + IRQF_SHARED, 0, &max77686_irq_chip,
>> + &max77686->irq_data);
>> + if (ret != 0) {
>> + dev_err(&i2c->dev, "failed to add PMIC irq chip: %d\n", ret);
>> + goto err_unregister_i2c;
>> + }
>> + ret = regmap_add_irq_chip(max77686->rtc_regmap, max77686->irq,
>> + IRQF_TRIGGER_FALLING | IRQF_ONESHOT |
>> + IRQF_SHARED, 0, &max77686_rtc_irq_chip,
>> + &max77686->rtc_irq_data);
>> + if (ret != 0) {
>> + dev_err(&i2c->dev, "failed to add RTC irq chip: %d\n", ret);
>> + goto err_del_irqc;
>> + }
>>
>> ret = mfd_add_devices(max77686->dev, -1, max77686_devs,
>> ARRAY_SIZE(max77686_devs), NULL, 0, NULL);
>> if (ret < 0) {
>> - mfd_remove_devices(max77686->dev);
>> - i2c_unregister_device(max77686->rtc);
>> + dev_err(&i2c->dev, "failed to add MFD devices: %d\n", ret);
>> + goto err_del_rtc_irqc;
>> }
>>
>> + return 0;
>> +
>> +err_del_rtc_irqc:
>> + regmap_del_irq_chip(max77686->irq, max77686->rtc_irq_data);
>> +err_del_irqc:
>> + regmap_del_irq_chip(max77686->irq, max77686->irq_data);
>
> I would imagine you either don't need these regmap_del_irq_chip() here
> or that you _do_ need them in max77686_i2c_remove().
>
> ...from looking at other drivers I think the answer is to add them to
> max77686_i2c_remove().
>
>

Yes, I forgot to do the regmap irqchip cleanup in the driver .remove function.
The same issue is in the max77802 driver btw so I'll fix both on v3.

Thanks a lot for pointing this out.

>> diff --git a/include/linux/mfd/max77686-private.h b/include/linux/mfd/max77686-private.h
>> index 8c75a9c..3a810b1 100644
>> --- a/include/linux/mfd/max77686-private.h
>> +++ b/include/linux/mfd/max77686-private.h
>> @@ -205,7 +205,7 @@ enum max77686_irq {
>> MAX77686_PMICIRQ_140C,
>> MAX77686_PMICIRQ_120C,
>>
>> - MAX77686_RTCIRQ_RTC60S,
>> + MAX77686_RTCIRQ_RTC60S = 0,
>> MAX77686_RTCIRQ_RTCA1,
>> MAX77686_RTCIRQ_RTCA2,
>> MAX77686_RTCIRQ_SMPL,
>> @@ -215,6 +215,25 @@ enum max77686_irq {
>> MAX77686_IRQ_NR,
>
> Maybe remove MAX77686_IRQ_NR which no longer makes any sense now that
> you start over at 0 partway through.
>
>

Right, it is not needed anymore. I'll remove it when re-spin.

> Overall this looks good to me, so once nits above are fixed feel to
> add my Reviewed-by. I've also built and booted this patch on
> exynos5250-snow and tested that the RTC wakealarm fires and can even
> wake the system up (with some additional work that I'll email you
> about).
>
> Reviewed-by: Doug Anderson <dianders@xxxxxxxxxxxx>
> Tested-by: Doug Anderson <dianders@xxxxxxxxxxxx>
>

Great! thanks a lot for testing.

Best regards,
Javier
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/