Messages in this thread | | | Date | Wed, 25 May 2011 13:47:39 -0700 | From | "Paul E. McKenney" <> | Subject | Re: [tip:core/rcu] Revert "rcu: Decrease memory-barrier usage based on semi-formal proof" |
| |
On Wed, May 25, 2011 at 09:27:42AM +0200, Ingo Molnar wrote: > > * Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote: > > > On Tue, May 24, 2011 at 05:10:11PM -0700, Yinghai Lu wrote: > > > On 05/24/2011 02:23 PM, Yinghai Lu wrote: > > > > On 05/23/2011 06:35 PM, Paul E. McKenney wrote: > > > >> On Mon, May 23, 2011 at 06:26:23PM -0700, Yinghai Lu wrote: > > > >>> On 05/23/2011 06:18 PM, Paul E. McKenney wrote: > > > >>> > > > >>>> OK, so it looks like I need to get this out of the way in order to track > > > >>>> down the delays. Or does reverting PeterZ's patch get you a stable > > > >>>> system, but with the longish delays in memory_dev_init()? If the latter, > > > >>>> it might be more productive to handle the two problems separately. > > > >>>> > > > >>>> For whatever it is worth, I do see about 5% increase in grace-period > > > >>>> duration when switching to kthreads. This is acceptable -- your > > > >>>> 30x increase clearly is completely unacceptable and must be fixed. > > > >>>> Other than that, the main thing that affects grace period duration is > > > >>>> the setting of CONFIG_HZ -- the smaller the HZ value, the longer the > > > >>>> grace-period duration. > > > >>> > > > >>> for my 1024g system when memory hotadd is enabled in kernel config: > > > >>> 1. current linus tree + tip tree: memory_dev_init will take about 100s. > > > >>> 2. current linus tree + tip tree + your tree - Peterz patch: > > > >>> a. on fedora 14 gcc: will cost about 4s: like old times > > > >>> b. on opensuse 11.3 gcc: will cost about 10s. > > > >> > > > >> So some patch in my tree that is not yet in tip makes things better? > > > >> > > > >> If so, could you please see which one? Maybe that would give me a hint > > > >> that could make things better on opensuse 11.3 as well. > > > > > > > > today's tip: > > > > > > > > [ 31.795597] cpu_dev_init done > > > > [ 40.930202] memory_dev_init done > > > > > > > > > > another boot from tip got: > > > > > > [ 35.211927] cpu_dev_init done > > > [ 136.053698] memory_dev_init done > > > > > > wonder if you can have clean revert for > > > > > > commit a26ac2455ffcf3be5c6ef92bc6df7182700f2114 > > > > Author: Paul E. McKenney <paul.mckenney@linaro.org> > > > > Date: Wed Jan 12 14:10:23 2011 -0800 > > > > > > > > rcu: move TREE_RCU from softirq to kthread > > > > > > > > If RCU priority boosting is to be meaningful, callback invocation must > > > > be boosted in addition to preempted RCU readers. Otherwise, in presence > > > > of CPU real-time threads, the grace period ends, but the callbacks don't > > > > get invoked. If the callbacks don't get invoked, the associated memory > > > > doesn't get freed, so the system is still subject to OOM. > > > > > > > > But it is not reasonable to priority-boost RCU_SOFTIRQ, so this commit > > > > moves the callback invocations to a kthread, which can be boosted easily. > > > > > > > > Also add comments and properly synchronized all accesses to > > > > rcu_cpu_kthread_task, as suggested by Lai Jiangshan. > > > > > > > > Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> > > > > Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> > > > > Reviewed-by: Josh Triplett <josh@joshtriplett.org> > > > > There is a new branch yinghai.2011.05.24a on: > > > > git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-2.6-rcu.git > > > > Or will be as soon as kernel.org updates its mirrors. > > > > I am not sure I could call this "clean", but it does revert that commit > > and 11 of the subsequent commits that depend on it. It does build, > > and I will test it once my currently running tests complete. > > Given that this is about a 1-2 minute delays with 1 *terabyte* of RAM, the per > gigabyte delay is like 60-120 msecs, right? > > So it's not a regression we are absolutely forced to address via a quick > revert, debugging it would be nicer. There's something we don't understand and > that's arguably worse than having unresolved non-fatal bugs :-)
And my attempted revert results in test failures in any case. :-(
> We already fixed the worst problem via a revert, the semi-hang: so i don't > think there's pressure to do other reverts - other than for diagnostic > purposes, of course.
Given that I have to debug in any case, I am happier debugging in the forward direction rather than in the backwards direction. ;-)
Thanx, Paul
| |