GDB 5.0 2000-03-05

[H . J . Lu]

On Tue, Mar 07, 2000 at 11:06:49PM -0500, Jim Kingdon wrote:
> > I like this approach.  Then we can simply enable thread_db when we've
> > fixed all critical bugs.
> 
> Agreed.  There just isn't time to get thread_db working for 5.0
> (IMHO).  I'd be interested in hearing reports of how much disabling
> thread_db helps, though (e.g. with mozilla).  The spurious SIGTRAP bug
> is not specific to thread_db but we have at least one report (HJ's)
> that it is worse with thread_db.
> 
> > But the patch looks wrong to me.  Seems to *enable* thread_db by
> > default, end what's up with LIN_CTHREAD_CFLAGS?
> 
> The patch looks OK to me.  If you don't specify --enable-lin-thread
> then thread_db doesn't get set, so LIN_THREAD gets set to empty.  I'm
> not enough of a configure.in person to know whether there is an easier
> way.  Agreed about LIN_CTHREAD_CFLAGS though, I don't see that being
> used anywhere.

I put it there just in case that thread_db requires some changes
in source code which break the old one.


H.J.