On Wed, Jan 18, 2006 at 10:57:47AM +1100, Keith Owens wrote:
> "Paul E. McKenney" (on Tue, 17 Jan 2006 09:26:17 -0800) wrote:
[ . . .]
> >One other thing -- given that you are requiring that the read side
> >have preemption disabled, another update-side option would be to
> >use synchronize_sched() to wait for all preemption-disabled code
> >segments to complete. This would allow you to remove the per-CPU
> >reference counters from the read side, but would require that the
> >update side either (1) be able to block or (2) be able to defer
> >the cleanup to process context.
>
> Originally I looked at that code but the comment scared me off.
> synchronize_sched() maps to synchronize_rcu() and the comment says that
> this only protects against rcu_read_lock(), not against preempt_disable().
Good point -- in mainline kernels (but not in some realtime
variants), the two guarantees happen to be one and the same.
But the comment certainly does not make this clear.
> /**
> * synchronize_sched - block until all CPUs have exited any non-preemptive
> * kernel code sequences.
> *
> * This means that all preempt_disable code sequences, including NMI and
> * hardware-interrupt handlers, in progress on entry will have completed
> * before this primitive returns. However, this does not guarantee that
> * softirq handlers will have completed, since in some kernels
> *
> * This primitive provides the guarantees made by the (deprecated)
> * synchronize_kernel() API. In contrast, synchronize_rcu() only
> * guarantees that rcu_read_lock() sections will have completed.
> */
> #define synchronize_sched() synchronize_rcu()
Does the following change help?
Thanx, Paul
diff -urpNa -X dontdiff linux-2.6.15/include/linux/rcupdate.h
linux-2.6.15-RCUcomment/include/linux/rcupdate.h
--- linux-2.6.15/include/linux/rcupdate.h 2006-01-02 19:21:10.000000000
-0800
+++ linux-2.6.15-RCUcomment/include/linux/rcupdate.h 2006-01-17
18:48:33.000000000 -0800
@@ -265,11 +265,14 @@ static inline int rcu_pending(int cpu)
* This means that all preempt_disable code sequences, including NMI and
* hardware-interrupt handlers, in progress on entry will have completed
* before this primitive returns. However, this does not guarantee that
- * softirq handlers will have completed, since in some kernels
+ * softirq handlers will have completed, since in some kernels, these
+ * handlers can run in process context, and can block.
*
* This primitive provides the guarantees made by the (deprecated)
* synchronize_kernel() API. In contrast, synchronize_rcu() only
* guarantees that rcu_read_lock() sections will have completed.
+ * In "classic RCU", these two guarantees happen to be one and
+ * the same, but can differ in realtime RCU implementations.
*/
#define synchronize_sched() synchronize_rcu()
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech
