Re: Reproducable kernel crash

On Tue, Feb 27, 2024 at 01:29:03PM +0100, Martin Husemann wrote:
> basically:
> 
> 	do_lwp_create( ...., &l2, ...);
> 	lwp_exit(l2);
> 
> There is a pmap_activate() call in there that makes l2's pm_pcbs have
> two entries, and then during lwp_exit only the active one gets removed.

I think this comes from vax/vm_machdep.c:113 where the new lwp gets
forcefully activated (which is wrong, isnt't it?)


	pcb2->pcb_pm = NULL;
	pmap_activate(l2);

I don't see any reason for this pmap_activate(), it would be done later
if needed, and never in our sample at hand.

Martin