Comment 5 for bug 1680386

------- Comment From <email address hidden> 2017-11-07 08:52 EDT-------
(In reply to comment #7)
> Hi,
> in this cae guest message are not important.
> Also I already said that it is non fatal - so yes it works, still it is
> wrong.
>
> Either fix or remove the code whatever is more appropriate.
> Please check the code that I referred to in qemu.
>
> hw/ppc/spapr_pci.c:775
> 774 /* Construct and read from host device tree the loc-code */
> 775 path = g_strdup_printf("/proc/device-tree%s/ibm,loc-code", buf);
> 776 g_free(buf);
>
> I'm rather convinced there should be an extra "/" in there.

(In reply to comment #7)
> Hi,
> in this cae guest message are not important.
> Also I already said that it is non fatal - so yes it works, still it is
> wrong.
>
> Either fix or remove the code whatever is more appropriate.
> Please check the code that I referred to in qemu.
>
> hw/ppc/spapr_pci.c:775
> 774 /* Construct and read from host device tree the loc-code */
> 775 path = g_strdup_printf("/proc/device-tree%s/ibm,loc-code", buf);
> 776 g_free(buf);
>
> I'm rather convinced there should be an extra "/" in there.

Thanks Paelzer, We had submitted the patch based upon your suggestion and awaiting response.
.
Patch which we submitted will be fixing 1. device-tree path string to use /proc/device-tree/%s/ibm,loc-code.