'make -j16 universe' gives SIReset

[Marius Strobl]

On Wed, Jun 29, 2011 at 12:54:33PM +1000, Peter Jeremy wrote:
> On 2011-Jun-22 12:05:24 +0200, Marius Strobl <marius at alchemy.franken.de> wrote:
> >Okay, given that it considerably improves the situation though I
> >suspect that the problem is that we instantly begin to fault on
> >kernel mappings once we flush all unlocked TLB entries in order
> >to get rid of the user mappings, which in case of cpu_switch()
> >still is covered by sched_lock. That would mean that we should use
> >a fine grained approach instead as the current one doesn't behave/
> >scale well even if sched_lock wasn't be (ab)used here. Could you
> >please give the following patch a try on top of what you already
> >have?
> >http://people.freebsd.org/~marius/sparc64_flush_user_no_sledgehammer.diff
> 
> My V890 has been running "make -j32 buildworld" in a loop for a
> week now without problems so I think that was the problem.
> 

Okay, fine, thanks for testing! Just to recap, apart from the above
patch you're running with the one-line patch for exception.S which
turns the sir in intr_vector into a retry and the patch for pmap.c
reducing the lock coverage in pmap_activate() (committed as r223347).
Have you additionally altered the DCR configuration in cheetah_init()?

Marius