Re: 2.6.23-rc2-mm1: rcutorture xtime usage

[Paul E. McKenney]

On Fri, Aug 10, 2007 at 01:30:55PM -0700, Paul E. McKenney wrote:
> On Fri, Aug 10, 2007 at 10:12:12AM -0700, Andrew Morton wrote:
> > On Fri, 10 Aug 2007 08:12:08 -0700 "Paul E. McKenney" <paulmck@xxxxxxxxxxxxxxxxxx> wrote:
> > 
> > > > One used to use sched_clock() for this, then get frowned at.  Now we
> > > > have cpu_clock()...
> > > 
> > > Hmmm...  And cpu_clock() is not in 2.6.22, so must appear in some later
> > > release.  Which means that the rate of API change in this area is a
> > > bit high, so I should avoid it like the plague.
> > 
> > eh, it's been there for weeks.  It is dust-encrusted.
> > 
> > >  Therefore, I should
> > > look for some other convenient source of entropy.
> > > 
> > > One convenient source would the per-CPU statistics that rcutorture
> > > maintains.  Of course, a given CPU's RNG is nearly in lock-step with
> > > its own statistics, but not with the adjacent CPU's statistics...
> > > 
> > > I will send a patch.
> > 
> > Please use cpu_clock().  It ain't going away.
> 
> D'accord...

Errmmm...  No joy.

	ERROR: "cpu_clock" [kernel/rcutorture.ko] undefined!

Turns out that cpu_clock also ain't exported, and rcutorture.c is
a module.  Would adding an EXPORT_SYMBOL_GPL() as in the patch below
be acceptable?

If not, I have a tested patch to rcutorture.c that leverages statistical
counters.  Your choice.

						Thanx, Paul

Add an EXPORT_SYMBOL_GPL() for cpu_clock() and make rcutorture.c use it.
Compiles, but not yet tested.

Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>
---

 rcutorture.c |    8 ++------
 sched.c      |    2 ++
 2 files changed, 4 insertions(+), 6 deletions(-)

diff -urpNa -X dontdiff linux-2.6.23-rc2/kernel/rcutorture.c linux-2.6.23-rc2-rcutorturesched/kernel/rcutorture.c
--- linux-2.6.23-rc2/kernel/rcutorture.c	2007-08-03 19:49:55.000000000 -0700
+++ linux-2.6.23-rc2-rcutorturesched/kernel/rcutorture.c	2007-08-10 17:15:22.000000000 -0700
@@ -42,7 +42,6 @@
 #include <linux/notifier.h>
 #include <linux/freezer.h>
 #include <linux/cpu.h>
-#include <linux/random.h>
 #include <linux/delay.h>
 #include <linux/byteorder/swabb.h>
 #include <linux/stat.h>
@@ -166,16 +165,13 @@ struct rcu_random_state {
 
 /*
  * Crude but fast random-number generator.  Uses a linear congruential
- * generator, with occasional help from get_random_bytes().
+ * generator, with occasional help from cpu_clock().
  */
 static unsigned long
 rcu_random(struct rcu_random_state *rrsp)
 {
-	long refresh;
-
 	if (--rrsp->rrs_count < 0) {
-		get_random_bytes(&refresh, sizeof(refresh));
-		rrsp->rrs_state += refresh;
+		rrsp->rrs_state += (unsigned long)cpu_clock(smp_processor_id());
 		rrsp->rrs_count = RCU_RANDOM_REFRESH;
 	}
 	rrsp->rrs_state = rrsp->rrs_state * RCU_RANDOM_MULT + RCU_RANDOM_ADD;
diff -urpNa -X dontdiff linux-2.6.23-rc2/kernel/sched.c linux-2.6.23-rc2-rcutorturesched/kernel/sched.c
--- linux-2.6.23-rc2/kernel/sched.c	2007-08-03 19:49:55.000000000 -0700
+++ linux-2.6.23-rc2-rcutorturesched/kernel/sched.c	2007-08-10 17:22:57.000000000 -0700
@@ -394,6 +394,8 @@ unsigned long long cpu_clock(int cpu)
 	return now;
 }
 
+EXPORT_SYMBOL_GPL(cpu_clock);
+
 #ifdef CONFIG_FAIR_GROUP_SCHED
 /* Change a task's ->cfs_rq if it moves across CPUs */
 static inline void set_task_cfs_rq(struct task_struct *p)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/