On Mon, Jan 01, 2001 at 11:45:32AM +0100, Jaromír Dolecek wrote:
> Frank van der Linden wrote:
> > There is currently no better solution. Several applications check for this,
> > and to provide good emulation, uname() should provide these values.
> > I don't give a flying f*ck about the argument "but I want netscape
> > to return 'NetBSD'". If you want something cosmetic rather than having
> > some binaries actually *work at all*, you're out of your mind.
> 
> OK, binaries using glibc linuxthreads don't seem to work after this change
> (checked with realplay).  This is not a big loose however, as there is
> still some emulation problem with threaded application which is
> not tracked down yet - majority of the threaded apps stopped working after
> a while. Since I need to track that down and don't feel like
> implementing Linux realtime signals right now, I'll use lower
> 'release' number for my tree. Not a big deal :)

That seems to be an indication that the "execvee()" solution would be
best to make version handling for Linux emulation more flexible. Well,
second to fixing the emulation of course :-)

For binaries that use 'clone', currently 2.0.x seems appropriate,
because we don't fully support it. We do support all other system
calls up 'til 2.4, though.. I'd like to get this linux clone() problem
fixed.

- Frank