what if there's no __NR_oldmount nor __NR_mount?

[Andreas Jaeger]

Alexandre Oliva <aoliva@redhat.com> writes:

> On Mar 14, 2003, Andreas Jaeger <aj@suse.de> wrote:
>
>> Alexandre Oliva <aoliva@redhat.com> writes:
>>> This patch fixes linux/umount.S such that it doesn't fail to assemble
>>> if an old-style umount syscall is not available at all.  Ok?
>
>> In what situation would this happen?
>
> Building glibc for n32 or n64.  There's only __NR_umount2, no
> __NR_umount or __NR_oldumount.

I see umount2 now.

>> If neither of these is available, the function has to be implemented
>> anyway as a dummy returning ENOSYS together with a stub-warning.
>
> Err...  Why?  I mean, is there any code that actually calls it?

User-level programs might call it.

> Anyway, I realize x86_64 and hppa are in similar situations, and they
> have an umount.c that should do for mips64 as well.  I'll submit this
> change along with some other cvs adds for linux/mips/mips64.

Yes, please do it in a similar way - or just include the hppa version,

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.inka.de
    http://www.suse.de/~aj