sched/fair: Fix -Wunused-but-set-variable warnings
authorQian Cai <cai@lca.pw>
Tue, 20 Aug 2019 18:40:55 +0000 (14:40 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 12 Nov 2019 18:18:29 +0000 (19:18 +0100)
commit 763a9ec06c409dcde2a761aac4bb83ff3938e0b3 upstream.

Commit:

   de53fd7aedb1 ("sched/fair: Fix low cpu usage with high throttling by removing expiration of cpu-local slices")

introduced a few compilation warnings:

  kernel/sched/fair.c: In function '__refill_cfs_bandwidth_runtime':
  kernel/sched/fair.c:4365:6: warning: variable 'now' set but not used [-Wunused-but-set-variable]
  kernel/sched/fair.c: In function 'start_cfs_bandwidth':
  kernel/sched/fair.c:4992:6: warning: variable 'overrun' set but not used [-Wunused-but-set-variable]

Also, __refill_cfs_bandwidth_runtime() does no longer update the
expiration time, so fix the comments accordingly.

Signed-off-by: Qian Cai <cai@lca.pw>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Reviewed-by: Dave Chiluk <chiluk+linux@indeed.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: pauld@redhat.com
Fixes: de53fd7aedb1 ("sched/fair: Fix low cpu usage with high throttling by removing expiration of cpu-local slices")
Link: https://lkml.kernel.org/r/1566326455-8038-1-git-send-email-cai@lca.pw
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
kernel/sched/fair.c

index d5c032ec193d6c8e11872e445ca14150cff4d958..feeb52880d353397ac16a76001d2bb686e378112 100644 (file)
@@ -4091,21 +4091,16 @@ static inline u64 sched_cfs_bandwidth_slice(void)
 }
 
 /*
- * Replenish runtime according to assigned quota and update expiration time.
- * We use sched_clock_cpu directly instead of rq->clock to avoid adding
- * additional synchronization around rq->lock.
+ * Replenish runtime according to assigned quota. We use sched_clock_cpu
+ * directly instead of rq->clock to avoid adding additional synchronization
+ * around rq->lock.
  *
  * requires cfs_b->lock
  */
 void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b)
 {
-       u64 now;
-
-       if (cfs_b->quota == RUNTIME_INF)
-               return;
-
-       now = sched_clock_cpu(smp_processor_id());
-       cfs_b->runtime = cfs_b->quota;
+       if (cfs_b->quota != RUNTIME_INF)
+               cfs_b->runtime = cfs_b->quota;
 }
 
 static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg)