Re: [Qemu-devel] [PATCH V12 5/9] Revert "pci: don't export an internal function"

[Anthony PERARD <anthony.perard@citrix.com>]

On 14/06/12 16:30, Michael S. Tsirkin wrote:
>> >  Is this patch would be a good common property?
>> >
>> >  It's probably close to the patch you send earlier Jan.
>> >
>> >  If it's good, I'll resend my passthrough patch series with this patch.
>> >
>> >
>> >  diff --git a/hw/qdev-properties.c b/hw/qdev-properties.c
>> >  index b7b5597..9c7a271 100644
>> >  --- a/hw/qdev-properties.c
>> >  +++ b/hw/qdev-properties.c
>> >  @@ -928,6 +928,104 @@ PropertyInfo qdev_prop_blocksize = {
>> >        .max   = 65024,
>> >    };
>> >
>> >  +/* --- pci host address --- */
>> >  +
>> >  +static void get_pci_host_devaddr(Object *obj, Visitor *v, void *opaque,
>> >  +                                 const char *name, Error **errp)
>> >  +{
>> >  +    DeviceState *dev = DEVICE(obj);
>> >  +    Property *prop = opaque;
>> >  +    PCIHostDeviceAddress *addr = qdev_get_prop_ptr(dev, prop);
>> >  +    char buffer[4 + 2 + 2 + 2 + 1];
> One trick is this:
>      char buffer[] = "XXXX:XX:XX.X";

I love this trick, even if it use a bit of space in the binary, I think. 
I will use it.

> And comparing with the above, your buffer looks too short, doesn't it?

Yes, I think I forgot to count the ':'s and '.'.

>> >  +    char *p = buffer;
>> >  +
>> >  +    snprintf(buffer, sizeof(buffer), "%04x:%02x:%02x.%d",
>> >  +             addr->domain, addr->bus, addr->slot, addr->function);
> A good idea might be to check the return code. assert(ret<= sizeof buf).

I will.

Thanks,

-- 
Anthony PERARD