diff options
| author | Ben Segall <bsegall@google.com> | 2013-10-16 11:16:17 -0700 |
|---|---|---|
| committer | Ingo Molnar <mingo@kernel.org> | 2013-10-29 12:02:20 +0100 |
| commit | db06e78cc13d70f10877e0557becc88ab3ad2be8 (patch) | |
| tree | 867b99d17a5df49fc4c413940a99e6209e81e7a7 | |
| parent | 1ee14e6c8cddeeb8a490d7b54cd9016e4bb900b4 (diff) | |
| download | op-kernel-dev-db06e78cc13d70f10877e0557becc88ab3ad2be8.zip op-kernel-dev-db06e78cc13d70f10877e0557becc88ab3ad2be8.tar.gz | |
sched: Fix cfs_bandwidth misuse of hrtimer_expires_remaining
hrtimer_expires_remaining does not take internal hrtimer locks and thus
must be guarded against concurrent __hrtimer_start_range_ns (but
returning HRTIMER_RESTART is safe). Use cfs_b->lock to make it safe.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181617.22647.73829.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
| -rw-r--r-- | kernel/sched/fair.c | 14 |
1 files changed, 11 insertions, 3 deletions
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index ebd187f..897d977 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3285,7 +3285,13 @@ static const u64 min_bandwidth_expiration = 2 * NSEC_PER_MSEC; /* how long we wait to gather additional slack before distributing */ static const u64 cfs_bandwidth_slack_period = 5 * NSEC_PER_MSEC; -/* are we near the end of the current quota period? */ +/* + * Are we near the end of the current quota period? + * + * Requires cfs_b->lock for hrtimer_expires_remaining to be safe against the + * hrtimer base being cleared by __hrtimer_start_range_ns. In the case of + * migrate_hrtimers, base is never cleared, so we are fine. + */ static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire) { struct hrtimer *refresh_timer = &cfs_b->period_timer; @@ -3361,10 +3367,12 @@ static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b) u64 expires; /* confirm we're still not at a refresh boundary */ - if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) + raw_spin_lock(&cfs_b->lock); + if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) { + raw_spin_unlock(&cfs_b->lock); return; + } - raw_spin_lock(&cfs_b->lock); if (cfs_b->quota != RUNTIME_INF && cfs_b->runtime > slice) { runtime = cfs_b->runtime; cfs_b->runtime = 0; |
