Hi Loic,
>>> + idev->irq = platform_get_irq(pdev, 0);
>>> + if (idev->irq < 0) {
>>> + struct gpio_desc *host_wake;
>>> +
>>> + dev_err(&pdev->dev, "No IRQ, falling back to gpio-irq\n");
>>> +
>>> + host_wake = devm_gpiod_get_optional(&pdev->dev, "host-wake",
>>> + GPIOD_IN);
>>> + if (IS_ERR(host_wake)) {
>>> + dev_err(&pdev->dev, "Unable to retrieve IRQ\n");
>>> + goto no_irq;
>>> + }
>>> +
>>> + idev->irq = gpiod_to_irq(host_wake);
>>> + if (idev->irq < 0) {
>>> + dev_err(&pdev->dev, "No corresponding irq for gpio\n");
>>> + goto no_irq;
>>> + }
>>> + }
>>> +
>>> + err = devm_request_threaded_irq(&pdev->dev, idev->irq, NULL, intel_irq,
>>> + IRQF_ONESHOT, "bt-host-wake", idev);
>>> + if (err) {
>>> + dev_err(&pdev->dev, "unable to allocate irq\n");
>>> + return err;
>>> + }
>>> +
>>> + device_init_wakeup(&pdev->dev, true);
>>
>> do you really want to do this in probe? Meaning that as soon as we boot, we claim this interrupt. Isn't it better to do this when the line discipline gets attached.
>
>
> Is it ok to move the irq request/free in intel_set_power?
I would assume so.
Regards
Marcel
--
To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
