[tip:sched/core] sched/core: Validate rq_clock*() serialization

From: tip-bot for Peter Zijlstra
Date: Wed Jan 14 2015 - 09:03:44 EST


Commit-ID: cebde6d681aa45f96111cfcffc1544cf2a0454ff
Gitweb: http://git.kernel.org/tip/cebde6d681aa45f96111cfcffc1544cf2a0454ff
Author: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
AuthorDate: Mon, 5 Jan 2015 11:18:10 +0100
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Wed, 14 Jan 2015 13:34:19 +0100

sched/core: Validate rq_clock*() serialization

rq->clock{,_task} are serialized by rq->lock, verify this.

One immediate fail is the usage in scale_rt_capability, so 'annotate'
that for now, there's more 'funny' there. Maybe change rq->lock into a
raw_seqlock_t?

(Only 32-bit is affected)

Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Link: http://lkml.kernel.org/r/20150105103554.361872747@xxxxxxxxxxxxx
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: umgwanakikbuti@xxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
kernel/sched/fair.c | 2 +-
kernel/sched/sched.h | 7 +++++++
2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 2a0b302..50ff902 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5948,8 +5948,8 @@ static unsigned long scale_rt_capacity(int cpu)
*/
age_stamp = ACCESS_ONCE(rq->age_stamp);
avg = ACCESS_ONCE(rq->rt_avg);
+ delta = __rq_clock_broken(rq) - age_stamp;

- delta = rq_clock(rq) - age_stamp;
if (unlikely(delta < 0))
delta = 0;

diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 9a2a45c..bd23732 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -687,13 +687,20 @@ DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
#define cpu_curr(cpu) (cpu_rq(cpu)->curr)
#define raw_rq() raw_cpu_ptr(&runqueues)

+static inline u64 __rq_clock_broken(struct rq *rq)
+{
+ return ACCESS_ONCE(rq->clock);
+}
+
static inline u64 rq_clock(struct rq *rq)
{
+ lockdep_assert_held(&rq->lock);
return rq->clock;
}

static inline u64 rq_clock_task(struct rq *rq)
{
+ lockdep_assert_held(&rq->lock);
return rq->clock_task;
}

--
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/