[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20070803184449.GA2196@linux.vnet.ibm.com>
Date: Fri, 3 Aug 2007 11:44:49 -0700
From: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To: linux-rt-users@...r.kernel.org
Cc: linux-kernel@...r.kernel.org, mingo@...e.hu, dipankar@...ibm.com,
ego@...ibm.com, vatsa@...ibm.com, josht@...ux.vnet.ibm.com
Subject: [PATCH RFC -rt] Allow rcutorture to handle synchronize_sched()
The rcutorture module uses a random-number generator that periodically
calls get_random_bytes() to add entropy. This random-number generator
is invoked from within the RCU read-side critical section, which, in the
case of the synchronize_sched() variant of RCU, has preeemption disabled.
Unfortunately, get_random_bytes() ends up acquiring normal spinlocks,
which can block in -rt, resulting in very large numbers of "scheduling
while atomic" messages.
This patch takes a very crude approach, simply substituting the time
of day for get_random_bytes().
Signed-off-by: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
---
rcutorture.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff -urpNa -X dontdiff linux-2.6.22.1-rt4/kernel/rcutorture.c linux-2.6.22.1-rt4-rcutorturesched/kernel/rcutorture.c
--- linux-2.6.22.1-rt4/kernel/rcutorture.c 2007-07-21 16:58:22.000000000 -0700
+++ linux-2.6.22.1-rt4-rcutorturesched/kernel/rcutorture.c 2007-08-02 16:38:09.000000000 -0700
@@ -165,16 +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 the time-of-day 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 += xtime.tv_nsec;
rrsp->rrs_count = RCU_RANDOM_REFRESH;
}
rrsp->rrs_state = rrsp->rrs_state * RCU_RANDOM_MULT + RCU_RANDOM_ADD;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists