Gerd Hoffmann <kra...@redhat.com> writes:
> Hi,
>
>> and object properties. Most object properties are data that we already
>> have, except for the unique persistant object identifier. Windows
>
>> + case PROP_PERSISTENT_UNIQUE_OBJECT_IDENTIFIER:
>> + /* Should be persistant between sessions,
>> + * but using our objedt ID is "good enough"
>> + * for now */
>> + usb_mtp_add_u64(d, 0x0000000000000000);
>> + usb_mtp_add_u64(d, o->handle);
>> + break;
>
> Didn't I mention for an older version already that the inode number
> would be a better fit? The inode number is persistent and unique per
> filesystem. While being at it we might also add the device id, to
> handle the (unlikely) case that the mtp device spans multiple
> filesystems.
Seems Isaac is busy with school and stuff. I will pick these up and post
a follow up soon.
Bandan
> We stat() the file already and cache the result, so this is as simple as
> using o->stat.st_{dev,ino}. See "man 2 stat" for details.
>
> Other than that the patch series looks fine.
>
> Sorry for the delay. Now that vacation time, kvm forum and qemu 2.7
> freeze are over things should go faster for v4.
>
> cheers,
> Gerd
