Re: [Qemu-devel] When to use qemu/typedefs.h

[Thomas Huth <thuth@redhat.com>]

On 31/07/2019 08.37, Markus Armbruster wrote:
> Paolo Bonzini <pbonzini@redhat.com> writes:
> 
>> On 30/07/19 15:15, Eric Blake wrote:
>>>> We occasionally give up and use types directly rather than their typedef
>>>> names, flouting the coding style.  This patch does.  Trades messing with
>>>> qemu/typedefs.h for having to write 'struct' a few times.
>>
>> I think Markus made the right call here.  Using "struct Foo;" in headers
>> is a null price to pay if all you need is declaring a pointer-typed
>> field or parameter.
> 
> Eduardo posted a patch to HACKING to clarify this non-usage of typedef
> is okay.
> 
> Should we continue to mandate typedef names elsewhere?  It adds
> cognitive load: you have to decide where to put the typedef, and when
> not to use it.

IMHO we should get rid of mandating typedefs. They are causing too much
trouble - e.g. do you also remember the issues with duplicated typedefs
in certain compiler versions in the past? (these should be hopefully
gone now, but still...)

And many QEMU developers are also working on the Linux kernel, which
rather forbids typedefs. Having to switch your mind back and forth
whether to use typedefs or not is really annoying.

So if you ask me, stop mandating it! It's ok as optional feature in QEMU
for types that are used all over the place, but we really should not
enforce it for each and every struct anymore.

 Thomas