[PATCH v3 12/12] PCI: rockchip-ep: Handle PERST# signal in endpoint mode

Damien Le Moal dlemoal at kernel.org
Mon Oct 14 23:24:30 PDT 2024


On 10/12/24 21:31, Manivannan Sadhasivam wrote:
> On Fri, Oct 11, 2024 at 06:30:31PM +0900, Damien Le Moal wrote:
>> On 10/10/24 19:49, Manivannan Sadhasivam wrote:
>>>> +static int rockchip_pcie_ep_setup_irq(struct pci_epc *epc)
>>>> +{
>>>> +	struct rockchip_pcie_ep *ep = epc_get_drvdata(epc);
>>>> +	struct rockchip_pcie *rockchip = &ep->rockchip;
>>>> +	struct device *dev = rockchip->dev;
>>>> +	int ret;
>>>> +
>>>> +	if (!rockchip->ep_gpio)
>>>> +		return 0;
>>>> +
>>>> +	/* PCIe reset interrupt */
>>>> +	ep->perst_irq = gpiod_to_irq(rockchip->ep_gpio);
>>>> +	if (ep->perst_irq < 0) {
>>>> +		dev_err(dev, "No corresponding IRQ for PERST GPIO\n");
>>>> +		return ep->perst_irq;
>>>> +	}
>>>> +
>>>> +	ep->perst_asserted = true;
>>>
>>> How come?
>>
>> Yeah, a bit confusing. This is because the gpio active low / inactive high, so
>> as soon as we enable the IRQ, we are going to get one IRQ even though perst gpio
>> signal has not changed yet.
> 
> Which means you are looking for a wrong level! What is the polarity of the
> PERST# gpio in DT?

It is not defined in the default DT with the kernel. I added an overlay file to
define it together with the EP mode. And as I said above, the gpio is active
low. If I reverse that to active high, it does not work.


-- 
Damien Le Moal
Western Digital Research



More information about the Linux-rockchip mailing list