This is the mail archive of the rda@sources.redhat.com mailing list for the rda project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: NPTL work committed to jimb-rda-nptl-branch


On Thu, Dec 02, 2004 at 07:44:10PM -0500, Jim Blandy wrote:
> Daniel Jacobowitz <drow@false.org> writes:
> 
> > On Tue, Nov 23, 2004 at 01:12:57AM -0500, Jim Blandy wrote:
> > > + /* Under NPTL, LWP's simply disappear, without becoming a zombie or
> > > +    producing any wait status.  At the kernel level, we have no way of
> > > +    knowing that the LWP's PID is now free and may be reused ---
> > > +    perhaps by an entirely different program!  So we need to use the
> > > +    death events from libthread_db to help us keep our LWP table clean.
> > 
> > FYI, that should not be true.  I think there was at least one broken
> > Red Hat kernel which behaved this way, but nowadays the debugger should
> > see exit events for LWPs.
> 
> This code isn't using ptrace events, though.  This version is still
> trying to struggle along with waitpid and thread_db.

By "event" I meant an exit from waitpid.  If waitpid is really letting
an attached thread disappear without reporting exit status to the
debugger, then something is wrong.

-- 
Daniel Jacobowitz


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]