Re: [PATCH] PCI: don't UNSET resource even reassign fails

[Bjorn Helgaas <bhelgaas@google.com>]

On Mon, Jun 16, 2014 at 07:29:42PM +0800, Guo Chao wrote:
> Commit bd064f0a sets IORESOURCE_UNSET flag if resource can not be
> assgined. Part of these changes breaks resource assignment.
> 
> If resource is too big, PCI core assgins basic part first and
> extends (reassigns) it to include optional part (notably SR-IOV resources).
> In this case, failure is permitted and IORESOURCE_UNSET should be not set.
> Otherwise, these resources are never updated to hardware even they are
> already propery assigned and we will see errors like this:
> 
> 	pci 0003:00:00.0: can't enable device: BAR 15 [mem size 0x0c000000 64bit pref] not assigned
> 	pci 0003:00:00.0: Error enabling bridge (-22), continuing
> 
> Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com>
> ---
>  drivers/pci/setup-res.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c
> index caed1ce..1e1076a 100644
> --- a/drivers/pci/setup-res.c
> +++ b/drivers/pci/setup-res.c
> @@ -334,12 +334,14 @@ int pci_reassign_resource(struct pci_dev *dev, int resno, resource_size_t addsiz
>  	new_size = resource_size(res) + addsize;
>  	ret = _pci_assign_resource(dev, resno, new_size, min_align);
>  	if (!ret) {
> -		res->flags &= ~IORESOURCE_UNSET;
>  		res->flags &= ~IORESOURCE_STARTALIGN;
>  		dev_info(&dev->dev, "BAR %d: reassigned %pR\n", resno, res);
>  		if (resno < PCI_BRIDGE_RESOURCES)
>  			pci_update_resource(dev, resno);
>  	}
> +
> +	res->flags &= ~IORESOURCE_UNSET;

This relies on the assumption that the resource was already assigned when
we entered pci_reassign_resource().  That *might* always be true, but it
takes a lot of code analysis to prove it.  I think I'd be OK with restoring
the original value of the IORESOURCE_UNSET bit.

>  	return ret;
>  }
>  
> -- 
> 1.8.3.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pci" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html