ObjC PR/6328

[Nicola Pero]

Hi, I examined your bug report on the gcc GNATS database, PR/6328,
concerning `+initialize being called multiple times'.

I couldn't reproduce your problem :-(

The code sending +initialize hasn't been changed in the past couple of
years at least, but the code looked ok to me.

I think I remember having had a problem with multiple +initialize being
called at some point in the past, and the whole gnustep has [not sure if
it's a historical relict or still needed!] every single +initialize
manually protected from multiple executions - so I strongly suspect there
might be a bug/problem in some obscure cases (or at least there was in the
past), and I'd like to fix it (or make sure it's no longer there), but no
matter how hard I tried, I couldn't reproduce it now.  It plainly works,
and the problem is that looking at the libobjc code it's clear why it
works too :-)

Do you have a testcase ?  That would be a great help.

Maybe something is messed up when many modules are loaded ... but even
with gnustep, it looks like it's normally working all right.

I would really appreciate a testcase.

Thanks.