[Buildroot] Buildroot fails on powerpc with busybox login.c

[Buildroot] Buildroot fails on powerpc with busybox login.c

[Michael Boutte]

Hi,
Ever since approximately version 16076 the initial make of buildroot 
fails with errors in the busybox/loginutils/login.c as shown in the snip 
below.
  CC loginutils/login.o
/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c: 
In function 'login_main':
/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:143: 
warning: implicit declaration of function 'read_or_build_utent'
/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:145: 
error: 'utent' undeclared (first use in this function)
/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:145: 
error: (Each undeclared identifier is reported only once
/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:145: 
error: for each function it appears in.)
/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:204: 
warning: implicit declaration of function 'write_utent'
make[2]: *** 
[/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.o] 
Error 1
make[1]: *** [_all] Error 2
make[1]: Leaving directory 
`/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox'
make: *** 
[/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/busybox] 
Error 2

Mike

[Buildroot] Buildroot fails on powerpc with busybox login.c

[Bernhard Fischer]

On Sun, Sep 10, 2006 at 09:34:47AM -0700, Michael Boutte wrote:
>Hi,
>Ever since approximately version 16076 the initial make of buildroot 

Sounds like vda's r16075 change to me.

>fails with errors in the busybox/loginutils/login.c as shown in the snip 
>below.
>  CC loginutils/login.o
>/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c: 
>In function 'login_main':
>/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:143: 
>warning: implicit declaration of function 'read_or_build_utent'


r16112 has a partial fix to make it build again. We (e.g. vda ;) should
provide empty bodies for read_or_build_utent() and write_utent() if UTMP
support is disabled. This can go hand in hand with removing the
pointless forward decls for said functions.

thanks for pointing this out,
Bernhard

[Buildroot] Buildroot fails on powerpc with busybox login.c

[Rob Landley]

On Wednesday 13 September 2006 11:44 am, Bernhard Fischer wrote:
> On Sun, Sep 10, 2006 at 09:34:47AM -0700, Michael Boutte wrote:
> >Hi,
> >Ever since approximately version 16076 the initial make of buildroot 
> 
> Sounds like vda's r16075 change to me.
> 
> >fails with errors in the busybox/loginutils/login.c as shown in the snip 
> >below.
> >  CC loginutils/login.o
> 
>/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c: 
> >In function 'login_main':
> 
>/home/mike/projects/software/snip875/build-tools/uclibc-4/buildroot/build_powerpc/busybox/loginutils/login.c:143: 
> >warning: implicit declaration of function 'read_or_build_utent'
> 
> 
> r16112 has a partial fix to make it build again. We (e.g. vda ;) should
> provide empty bodies for read_or_build_utent() and write_utent() if UTMP
> support is disabled. This can go hand in hand with removing the
> pointless forward decls for said functions.

Stick "ALWAYS_INLINE" on 'em when you do, please.

Rob
-- 
Never bet against the cheap plastic solution.

[Buildroot] Buildroot fails on powerpc with busybox login.c

[Bernhard Fischer]

On Thu, Sep 14, 2006 at 07:00:12PM +0200, Denis Vlasenko wrote:
>On Wednesday 13 September 2006 17:44, Bernhard Fischer wrote:

>> r16112 has a partial fix to make it build again. We (e.g. vda ;) should
>> provide empty bodies for read_or_build_utent() and write_utent() if UTMP
>> support is disabled. This can go hand in hand with removing the
>> pointless forward decls for said functions.
>
>Done ;)

Thanks for taking care of this.
cheers,
Bernhard