Re: [PATCH 12/15] hyperv: move VMBus connection ids to uapi

[Paolo Bonzini]

On 09/01/2017 09:40, hpa@xxxxxxxxx wrote:
> On January 9, 2017 12:32:23 AM PST, Roman Kagan <rkagan@xxxxxxxxxxxxx> wrote:
>> On Mon, Jan 02, 2017 at 09:19:57AM +0100, Paolo Bonzini wrote:
>>> On 28/12/2016 18:09, Roman Kagan wrote:
>>>> Am I correct assuming that QEMU is currently the only user of
>>>> arch/x86/include/uapi/asm/hyperv.h?
>>>>
>>>> Then I think we're fine withdrawing it from uapi as a whole and
>> letting
>>>> QEMU pull it in through its header-harvesting scripts (as does now
>>>> anyway).  This would lift all licensing and longterm API stability
>>>> expectations.
>>>
>>> Actually, QEMU's header-harvesting scripts use uapi/ headers
>>> exclusively, since they are built on "make headers_install".
>>>
>>> The extra cleanups that QEMU does on top are to allow compilation of
>> the
>>> headers on non-Linux machines.  They don't really do much more than
>>> changing Linux (linux/types.h) integer types to the C99 (stdint.h)
>>> equivalents.
>>
>> Ouch, I stand corrected.
>>
>> So what should we do with it then?  I'm sorta lost...
>>
>> We certainly can give it up and live with a private copy of the
>> definitions in the QEMU tree but that doesn't sound optimal in any
>> sense.
> 
> Why do that through header mangling rather than typedef?

Because you are not suppose to typedef identifiers that start with "__",
and because it does do a few other ad-hoc changes:

        -e 's/<linux\/\([^>]*\)>/"standard-headers\/linux\/\1"/' \
        -e 's/__bitwise__//' \
        -e 's/__attribute__((packed))/QEMU_PACKED/' \
        -e 's/__inline__/inline/' \
        -e '/sys\/ioctl.h/d' \
        -e 's/SW_MAX/SW_MAX_/' \

Paolo