Re: [PATCH] Make rcutorture more vicious: add stutter feature

[Ingo Molnar <mingo@elte.hu>]

* Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote:

> Hello!
> 
> This patch takes a step towards making rcutorture more brutal by 
> allowing the test to be automatically periodically paused, with the 
> default being to run the test for five seconds then pause for five 
> seconds and repeat. This behavior can be controlled using a new 
> "stutter" module parameter, so that "stutter=0" gives the old default 
> behavior of running continuously.
> 
> Starting and stopping rcutorture more heavily stresses RCU's 
> interaction with the scheduler, as well as exercising more paths 
> through the grace-period detection code.

cool! Applied to tip/core/rcu.

> Note that the default to "shuffle_interval" has also been adjusted 
> from 5 seconds to 3 seconds to provide varying overlap with the 
> "stutter" interval.
> 
> I am still unable to provoke the failures that Alexey has been seeing, 
> even with this patch, but will be doing a few additional things to 
> beef up rcutorture.

neither am i able to see any rcu-preempt+rcu-torture failures in my 
testing, despite having added the patch below to tip/core/rcu as well to 
make any failures more apparent.

	Ingo

------------------>
commit 5af970a48f3ba0dd96a036b196c79dc923f28231
Author: Ingo Molnar <mingo@elte.hu>
Date:   Wed Jun 18 10:09:48 2008 +0200

    rcutorture: WARN_ON_ONCE(1) when detecting an error
    
    this makes it easier for automated tests to pick up such failures.

diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index 0334b6a..0ca7e9b 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -687,6 +687,7 @@ rcu_torture_printk(char *page)
 	if (i > 1) {
 		cnt += sprintf(&page[cnt], "!!! ");
 		atomic_inc(&n_rcu_torture_error);
+		WARN_ON_ONCE(1);
 	}
 	cnt += sprintf(&page[cnt], "Reader Pipe: ");
 	for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++)