FYI: Increase in schedlock.exp failures in LinuxThreads on an RH kernel

[Elena Zannoni]

Daniel Jacobowitz writes:
 > On Mon, Jul 28, 2003 at 12:48:17PM -0700, Roland McGrath wrote:
 > > > I noticed recently, after switching kernels (I'm temporarily running a
 > > > Debian system with a Red Hat kernel - don't ask) 
 > > 
 > > We won't ask.  But, as you might imagine, we do run the gdb test suite on
 > > our kernels, and we haven't seen this.  So you indeed must have hit upon a
 > > lucky combination of hardware and software between that RH kernel and your
 > > Debian userland that happens to bite.
 > 
 > Have you run a testsuite using LinuxThreads, but the NPTL-aware kernel? 
 > That should be enough to trigger it, I think.  It won't show up in
 > NPTL, because the SIGINT will be load balanced and only delivered to
 > one of the threads.
 > 

No, not usually. I can try though.

elena



 > > > Ingo/Roland - might want to export ShdPnd in the RH kernels...
 > > 
 > > We probably will.  I always used that patch when debugging the kernel
 > > signals code in the past.
 > 
 > Thanks.
 > 
 > -- 
 > Daniel Jacobowitz
 > MontaVista Software                         Debian GNU/Linux Developer