Re: [PATCH v12 1/7] PCI: endpoint: Add auxiliary resource query API

[Koichiro Den]

On Fri, Mar 27, 2026 at 12:31:09PM -0400, Frank Li wrote:
> On Fri, Mar 27, 2026 at 12:54:16PM +0900, Koichiro Den wrote:
> > Endpoint controller drivers may integrate auxiliary blocks (e.g. DMA
> > engines) whose register windows and descriptor memories metadata need to
> > be exposed to a remote peer. Endpoint function drivers need a generic
> > way to discover such resources without hard-coding controller-specific
> > helpers.
> >
> > Add pci_epc_count_aux_resources() / pci_epc_get_aux_resources() and the
> > corresponding pci_epc_ops callbacks. The count helper returns the number
> > of available resources via an output parameter, while the get helper
> > fills a caller-provided array of resources described by type, physical
> > address and size, plus type-specific metadata.
> >
> > Suggested-by: Manivannan Sadhasivam <mani@xxxxxxxxxx>
> > Signed-off-by: Koichiro Den <den@xxxxxxxxxxxxx>
> > ---
> > Changes in v12:
> >   - Add pci_epc_count_aux_resources() and pci_epc_ops.count_aux_resources
> >   - Use pci_epc_function_is_valid() instead of open-coding
> >
> >  drivers/pci/endpoint/pci-epc-core.c | 83 +++++++++++++++++++++++++++++
> >  include/linux/pci-epc.h             | 53 ++++++++++++++++++
> >  2 files changed, 136 insertions(+)
> >
> > diff --git a/drivers/pci/endpoint/pci-epc-core.c b/drivers/pci/endpoint/pci-epc-core.c
> > index 6c3c58185fc5..fded71a19d71 100644
> > --- a/drivers/pci/endpoint/pci-epc-core.c
> > +++ b/drivers/pci/endpoint/pci-epc-core.c
> > @@ -156,6 +156,89 @@ const struct pci_epc_features *pci_epc_get_features(struct pci_epc *epc,
> >  }
> >  EXPORT_SYMBOL_GPL(pci_epc_get_features);
> >
> > +/**
> > + * pci_epc_count_aux_resources() - count EPC-provided auxiliary resources
> > + * @epc: EPC device
> > + * @func_no: function number
> > + * @vfunc_no: virtual function number
> > + * @num_resources: returned number of auxiliary resources
> > + *
> > + * Some EPC backends integrate auxiliary blocks (e.g. DMA engines) whose control
> > + * registers and/or descriptor memories can be exposed to the host by mapping
> > + * them into BAR space. This helper queries how many such resources the backend
> > + * provides.
> > + *
> > + * Return: 0 on success, -EOPNOTSUPP if the backend does not support auxiliary
> > + * resource queries, or another -errno on failure.
> > + */
> > +int pci_epc_count_aux_resources(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
> > +				int *num_resources)
> 
> Most like such kinds APIs return count directly,
>   < 0,  means error.
>  >= 0,  means how many resource.

I chose to return 0 on success here based on Mani's earlier feedback on v11:
https://lore.kernel.org/linux-pci/lfbv6sjevz4xtjr6zt2hl3wap3bfwkvigqb6fdztqmutxgcvld@3t4x5gstr4nj/

I don't have a strong preference here, but since there seems to be no clear
policy/convention across the kernel, I would prefer to follow what is most
consistent and maintainable for the maintainers. Of course, I'm happy to switch
and respin if preferred.

Thanks for reviewing,
Koichiro

> 
> Frank