[Xenomai-core] enum rtdm_selecttype

[Xenomai-core] enum rtdm_selecttype

[Gilles Chanteperdrix]

Hi Jan,

when compiling xenomai v2.4.x for ARM with gcc 4.2.1, I get plenty of
warnings like:
xenomai-arm/kernel/xenomai/skins/posix/syscall.c:38:
xenomai-arm/include/xenomai/rtdm/rtdm_driver.h:206: warning: parameter has incomplete type

So it seems gcc 4.2.1 does not like the enum being used in function
prototypes without being defines. So, could we move the enum definition
before the functions using it ?

-- 


					    Gilles.

Re: [Xenomai-core] enum rtdm_selecttype

[Philippe Gerum]

Gilles Chanteperdrix wrote:
> Hi Jan,
> 
> when compiling xenomai v2.4.x for ARM with gcc 4.2.1, I get plenty of
> warnings like:
> xenomai-arm/kernel/xenomai/skins/posix/syscall.c:38:
> xenomai-arm/include/xenomai/rtdm/rtdm_driver.h:206: warning: parameter has incomplete type
> 
> So it seems gcc 4.2.1 does not like the enum being used in function
> prototypes without being defines. So, could we move the enum definition
> before the functions using it ?
> 

We can sort this out with a simple forward declaration.

-- 
Philippe.

Re: [Xenomai-core] enum rtdm_selecttype

[Gilles Chanteperdrix]

Philippe Gerum wrote:
 > Gilles Chanteperdrix wrote:
 > > Hi Jan,
 > > 
 > > when compiling xenomai v2.4.x for ARM with gcc 4.2.1, I get plenty of
 > > warnings like:
 > > xenomai-arm/kernel/xenomai/skins/posix/syscall.c:38:
 > > xenomai-arm/include/xenomai/rtdm/rtdm_driver.h:206: warning: parameter has incomplete type
 > > 
 > > So it seems gcc 4.2.1 does not like the enum being used in function
 > > prototypes without being defines. So, could we move the enum definition
 > > before the functions using it ?
 > > 
 > 
 > We can sort this out with a simple forward declaration.

That is the way it is done today, but gcc 4.2.1 seems not to like
it. Maybe recent C standards allow the compiler to pick the smallest type
for the enum as C++ does ?

-- 


					    Gilles.

Re: [Xenomai-core] enum rtdm_selecttype

[Philippe Gerum]

Gilles Chanteperdrix wrote:
> Philippe Gerum wrote:
>  > Gilles Chanteperdrix wrote:
>  > > Hi Jan,
>  > > 
>  > > when compiling xenomai v2.4.x for ARM with gcc 4.2.1, I get plenty of
>  > > warnings like:
>  > > xenomai-arm/kernel/xenomai/skins/posix/syscall.c:38:
>  > > xenomai-arm/include/xenomai/rtdm/rtdm_driver.h:206: warning: parameter has incomplete type
>  > > 
>  > > So it seems gcc 4.2.1 does not like the enum being used in function
>  > > prototypes without being defines. So, could we move the enum definition
>  > > before the functions using it ?
>  > > 
>  > 
>  > We can sort this out with a simple forward declaration.
> 
> That is the way it is done today, but gcc 4.2.1 seems not to like
> it. Maybe recent C standards allow the compiler to pick the smallest type
> for the enum as C++ does ?
> 

Or any integral type the implementation sees fit actually. Well, yes, maybe it
is indeed bits of the C++ standard dripping on the C implementation of GCC,
especially if forward decl of enums is not explicitly defined by the C standard.

-- 
Philippe.

Re: [Xenomai-core] enum rtdm_selecttype

[Gilles Chanteperdrix]

Philippe Gerum wrote:
 > Gilles Chanteperdrix wrote:
 > > Philippe Gerum wrote:
 > >  > Gilles Chanteperdrix wrote:
 > >  > > Hi Jan,
 > >  > > 
 > >  > > when compiling xenomai v2.4.x for ARM with gcc 4.2.1, I get plenty of
 > >  > > warnings like:
 > >  > > xenomai-arm/kernel/xenomai/skins/posix/syscall.c:38:
 > >  > > xenomai-arm/include/xenomai/rtdm/rtdm_driver.h:206: warning: parameter has incomplete type
 > >  > > 
 > >  > > So it seems gcc 4.2.1 does not like the enum being used in function
 > >  > > prototypes without being defines. So, could we move the enum definition
 > >  > > before the functions using it ?
 > >  > > 
 > >  > 
 > >  > We can sort this out with a simple forward declaration.
 > > 
 > > That is the way it is done today, but gcc 4.2.1 seems not to like
 > > it. Maybe recent C standards allow the compiler to pick the smallest type
 > > for the enum as C++ does ?
 > > 
 > 
 > Or any integral type the implementation sees fit actually. Well, yes, maybe it
 > is indeed bits of the C++ standard dripping on the C implementation of GCC,
 > especially if forward decl of enums is not explicitly defined by the C standard.

Actually it is gcc 3.4 which does not like the forward decl, not gcc 4.2

-- 


					    Gilles.

Re: [Xenomai-core] enum rtdm_selecttype

[Jan Kiszka]

[-- Attachment #1: Type: text/plain, Size: 1477 bytes --]

Gilles Chanteperdrix wrote:
> Philippe Gerum wrote:
>  > Gilles Chanteperdrix wrote:
>  > > Philippe Gerum wrote:
>  > >  > Gilles Chanteperdrix wrote:
>  > >  > > Hi Jan,
>  > >  > > 
>  > >  > > when compiling xenomai v2.4.x for ARM with gcc 4.2.1, I get plenty of
>  > >  > > warnings like:
>  > >  > > xenomai-arm/kernel/xenomai/skins/posix/syscall.c:38:
>  > >  > > xenomai-arm/include/xenomai/rtdm/rtdm_driver.h:206: warning: parameter has incomplete type
>  > >  > > 
>  > >  > > So it seems gcc 4.2.1 does not like the enum being used in function
>  > >  > > prototypes without being defines. So, could we move the enum definition
>  > >  > > before the functions using it ?
>  > >  > > 
>  > >  > 
>  > >  > We can sort this out with a simple forward declaration.
>  > > 
>  > > That is the way it is done today, but gcc 4.2.1 seems not to like
>  > > it. Maybe recent C standards allow the compiler to pick the smallest type
>  > > for the enum as C++ does ?
>  > > 
>  > 
>  > Or any integral type the implementation sees fit actually. Well, yes, maybe it
>  > is indeed bits of the C++ standard dripping on the C implementation of GCC,
>  > especially if forward decl of enums is not explicitly defined by the C standard.
> 
> Actually it is gcc 3.4 which does not like the forward decl, not gcc 4.2

I'm fine with moving it if that helps to make all compilers happy. Just
ensure that it remains in the right doxygen group.

Jan


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 254 bytes --]