Messages in this thread | | | Date | Wed, 12 Dec 2007 22:56:12 -0500 (EST) | From | Steven Rostedt <> | Subject | Re: [PATCH, RFC] hacks to allow -rt to run kernbench on POWER |
| |
On Mon, 29 Oct 2007, Paul E. McKenney wrote: > diff -urpNa -X dontdiff linux-2.6.23.1-rt4/mm/memory.c linux-2.6.23.1-rt4-fix/mm/memory.c > --- linux-2.6.23.1-rt4/mm/memory.c 2007-10-27 22:20:57.000000000 -0700 > +++ linux-2.6.23.1-rt4-fix/mm/memory.c 2007-10-28 15:40:36.000000000 -0700 > @@ -664,6 +664,7 @@ static unsigned long zap_pte_range(struc > int anon_rss = 0; > > pte = pte_offset_map_lock(mm, pmd, addr, &ptl); > + preempt_disable(); > arch_enter_lazy_mmu_mode(); > do { > pte_t ptent = *pte; > @@ -732,6 +733,7 @@ static unsigned long zap_pte_range(struc > > add_mm_rss(mm, file_rss, anon_rss); > arch_leave_lazy_mmu_mode(); > + preempt_enable(); > pte_unmap_unlock(pte - 1, ptl); > > return addr;
I'm pulling your patch for the above added code. Took me a few hours to find the culprit, but I was getting scheduling in atomic bugs. Turns out that this code you put "preempt_disable" in calls sleeping spinlocks.
Might want to run with DEBUG_PREEMPT.
Thanks,
-- Steve
| |