On 2/6/2015 8:50 AM, Romain Perier wrote: > Don't register interrupt handler before usb gadget is correctly initialized. > For some embedded platforms which don't have a usb-phy, it crashes the driver > because an interrupt is emitted with non-initialized hardware. > According to devm_request_irq documentation, an interrupt can be emitted > at any time once the interrupt is registered, so we have to care about driver > and hardware initialization. > > Signed-off-by: Romain Perier <romain.per...@gmail.com> > --- > > Changes for v2: fix typos in commit log > > drivers/usb/dwc2/platform.c | 17 +++++++++-------- > 1 file changed, 9 insertions(+), 8 deletions(-) > > diff --git a/drivers/usb/dwc2/platform.c b/drivers/usb/dwc2/platform.c > index ae095f0..b26cf8c 100644 > --- a/drivers/usb/dwc2/platform.c > +++ b/drivers/usb/dwc2/platform.c > @@ -196,14 +196,6 @@ static int dwc2_driver_probe(struct platform_device *dev) > return irq; > } > > - dev_dbg(hsotg->dev, "registering common handler for irq%d\n", > - irq); > - retval = devm_request_irq(hsotg->dev, irq, > - dwc2_handle_common_intr, IRQF_SHARED, > - dev_name(hsotg->dev), hsotg); > - if (retval) > - return retval; > - > res = platform_get_resource(dev, IORESOURCE_MEM, 0); > hsotg->regs = devm_ioremap_resource(&dev->dev, res); > if (IS_ERR(hsotg->regs)) > @@ -237,6 +229,15 @@ static int dwc2_driver_probe(struct platform_device *dev) > retval = dwc2_gadget_init(hsotg, irq); > if (retval) > return retval; > + > + dev_dbg(hsotg->dev, "registering common handler for irq%d\n", > + irq); > + retval = devm_request_irq(hsotg->dev, irq, > + dwc2_handle_common_intr, IRQF_SHARED, > + dev_name(hsotg->dev), hsotg); > + if (retval) > + return retval; > + > retval = dwc2_hcd_init(hsotg, irq, params); > if (retval) > return retval; >
This patch fails checkpatch.pl. Which interrupt is being emitted? You are only initializing the gadget side before registering the handler. Could it affect the host side as well, say if you config for HOST only? In that case none of the dwc2_gadget_init() will be done (as it is a no-op). Thanks, John -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
