RE: [PATCH V8 03/13] PCI: dwc: Parse Root Port nodes in dw_pcie_host_init()

[Sherry Sun]

> Subject: Re: [PATCH V8 03/13] PCI: dwc: Parse Root Port nodes in
> dw_pcie_host_init()
> 
> On Fri, Mar 13, 2026 at 10:08:13AM +0800, Sherry Sun wrote:
> > Add support for parsing Root Port child nodes in dw_pcie_host_init()
> > using pci_host_common_parse_ports(). This allows DWC-based drivers to
> > specify Root Port properties (like reset GPIOs) in individual Root
> > Port nodes rather than in the host bridge node.
> >
> > Signed-off-by: Sherry Sun <sherry.sun@xxxxxxx>
> > ---
> >  drivers/pci/controller/dwc/pcie-designware-host.c | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c
> > b/drivers/pci/controller/dwc/pcie-designware-host.c
> > index a74339982c24..9608f9fcd1b3 100644
> > --- a/drivers/pci/controller/dwc/pcie-designware-host.c
> > +++ b/drivers/pci/controller/dwc/pcie-designware-host.c
> > @@ -20,6 +20,7 @@
> >  #include <linux/platform_device.h>
> >
> >  #include "../../pci.h"
> > +#include "../pci-host-common.h"
> >  #include "pcie-designware.h"
> >
> >  static struct pci_ops dw_pcie_ops;
> > @@ -581,6 +582,13 @@ int dw_pcie_host_init(struct dw_pcie_rp *pp)
> >
> >  	pp->bridge = bridge;
> >
> > +	/* Parse Root Port nodes if present */
> > +	ret = pci_host_common_parse_ports(dev, bridge);
> > +	if (ret && ret != -ENOENT) {
> 
> Is there an incentive in the -ENOENT error code? Should the API just return 0
> if no ports are found?
> 

Hi Mani,

Since devm_fwnode_gpiod_get() is called within pci_host_common_parse_ports(),
devm_fwnode_gpiod_get() will return -ENOENT if no GPIO found. That means we
need to check and handle the -ENOENT error anyway, we can either handle it internally
in pci_host_common_parse_ports() or in the upper caller. 

If we hide -ENOENT inside pci_host_common_parse_ports() and always return 0, we
lose this distinction, the caller can no longer tell the difference between "not found" and
"found and parsed successfully". This might be useful that different callers might want to
handle differently.

The current pattern is also consistent with some other optional resource APIs in the
kernel (e.g., devm_regulator_get_optional()), where -ENOENT is returned and the caller
decides how to handle it.

Best Regards
Sherry