Subject: Re: kern/470: useless struct definition
To: None <email@example.com>
From: Stefan Grefen <firstname.lastname@example.org>
Date: 09/19/1994 11:02:04
In message <199409160211.TAA03130@sun-lamp.cs.berkeley.edu> email@example.com wrote:
> `J.T. Conklin' changed the state to `analyzed'.
> State-Changed-From-To: open-analyzed
> State-Changed-By: jtc
> State-Changed-When: Thu Sep 15 19:04:03 1994
> Examination of the NetBSD kernel source (sysv_msg.c) and the SunOS and
> OSF/1 <sys/msg.h> header files seems to indicate that the msginfo
> variable _is_ correct, but both it and its structure definition belong
> inside an #ifdef KERNEL. It's up to the kernel folks to take it from here.
> I don't know why you were having problems with multiply defined
> symbols, as the linker should have treated msginfo as a common.
I'm building a C++ shared-library and it looks like it doesn't go into the
common segment. (Well actually it's diffcult for C++ here to determine who is
going to call the constructor-code, if it is declared in X object files).
BTW. How can I make it easier for you? (e.g. how to attach a fix-suggesting?)
Normaly I always attach a fix to my PR's.
Stefan Grefen Convex Computer GmbH, Frankfurt, Germany
firstname.lastname@example.org Phone: +49-69-665270
fortune: "You can't make a program without broken egos."