[PATCH v7 0/3] Add support for PCI in AArch64

[Liviu Dudau]

Hi Sunil,

On Fri, May 16, 2014 at 11:33:04AM +0100, Sunil Kovvuri wrote:
> Hi Liviu,
> 
> I am using your ARM64 PCIe patches to write a PCIe host controller
> driver for our SOC. I am facing an issue with SR-IOV capable device.
> 
> Consider an PCI Express endpoint connected to a PCI Express
> Root Port.  The PCI Express endpoint provides PCI-SIG SR-IOV
> capabilities with a single physical function and a large number
> of virtual functions.  The Root Port contains a pci-pci bridge in
> between it and the SR-IOV device.
> 
> When the SR-IOV capable device's driver tries to enable sriov
> (pci_enable_sriov()) it fails to create/add PCI device for each
> virtual function reporting  "not enough MMIO resources for SR-IOV".
> 
> In sriov_enable() (drivers/pci/iov.c)
> 
>  296        for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
>  297                 bars |= (1 << (i + PCI_IOV_RESOURCES));
>  298                 res = dev->resource + PCI_IOV_RESOURCES + i;
>  299                 if (res->parent)
>  300                         nres++;
>  301         }
>  302         if (nres != iov->nres) {
>  303                 dev_err(&dev->dev, "not enough MMIO resources for
>  SR-IOV\n");
>  304                 return -ENOMEM;
>  305         }
> 
> Here its checking if physical function's IOV resource has a parent or not.
> Which is pci-pci bridge in this case. Otherwise it doesn't consider
> that resource.
> 
> Added below api to your patch.
> This will try to claim a resource while creating a PCI device which
> inturn sets 'res->parent'.
> 
> Let me know if this is okay.
> 
> diff --git a/arch/arm64/kernel/pci.c b/arch/arm64/kernel/pci.c
> index 9f29c9a..fbfb48f 100644
> --- a/arch/arm64/kernel/pci.c
> +++ b/arch/arm64/kernel/pci.c
> @@ -125,6 +125,21 @@ resource_size_t pcibios_align_resource(void
> *data, const struct resource *res,
>         return res->start;
>  }
> 
> +int pcibios_add_device(struct pci_dev *pdev)
> +{
> +        unsigned int i, type_mask = IORESOURCE_IO | IORESOURCE_MEM;
> +        struct resource *res;
> +
> +        for (i = 0; i < PCI_NUM_RESOURCES; i++) {
> +                res = &pdev->resource[i];
> +                if (res->parent || !(res->flags & type_mask))
> +                        continue;
> +                pci_claim_resource(pdev, i);
> +        }
> +
> 

I would like not to have to add your patch in this file as I am trying to
remove it entirely. I don't have an SR-IOV capable device in my setup so
I am not able to debug your problem, but could you have a go and try to
figure out why the SR-IOV resources do not get a parent associated with
when they get created?

Many thanks,
Liviu


-- 
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ¯\_(ツ)_/¯