On Wed, Jul 19, 2017 at 11:18 PM, Haiyang Zhang <haiyangz@xxxxxxxxxxxxx> wrote:
>> struct hv_fcopy_hdr {
>> __u32 operation;
>> - uuid_le service_id0; /* currently unused */
>> - uuid_le service_id1; /* currently unused */
>> + __u8 service_id0[16]; /* currently unused */
>> + __u8 service_id1[16]; /* currently unused */
>> } __attribute__((packed));
>
> Even user space may not have all guid_t operations, could you still
> put guid_t here for consistency?
I'm not sure how this code has been designed, though uuid_le followed
by guid_t are kernel internal types mistakenly put to UAPI in the
first place.
Consistency here is how we treat those types in user space, starting
from v4.13 uuid_be gone from that header. uuid_le is luckily will be
gone soon.
> This file is also used by our kernel
> mode code.
It will work still the same after this change.
--
With Best Regards,
Andy Shevchenko
