From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mon, 22 Jun 2015 22:52:04 +0000 (-0700)
Subject: Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel... 
X-Git-Tag: firefly_0821_release~176^2~1632
X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=23b7776290b10297fe2cae0fb5f166a4f2c68121;p=firefly-linux-kernel-4.4.55.git

Merge branch 'sched-core-for-linus' of git://git./linux/kernel/git/tip/tip

Pull scheduler updates from Ingo Molnar:
 "The main changes are:

   - lockless wakeup support for futexes and IPC message queues
     (Davidlohr Bueso, Peter Zijlstra)

   - Replace spinlocks with atomics in thread_group_cputimer(), to
     improve scalability (Jason Low)

   - NUMA balancing improvements (Rik van Riel)

   - SCHED_DEADLINE improvements (Wanpeng Li)

   - clean up and reorganize preemption helpers (Frederic Weisbecker)

   - decouple page fault disabling machinery from the preemption
     counter, to improve debuggability and robustness (David
     Hildenbrand)

   - SCHED_DEADLINE documentation updates (Luca Abeni)

   - topology CPU masks cleanups (Bartosz Golaszewski)

   - /proc/sched_debug improvements (Srikar Dronamraju)"

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (79 commits)
  sched/deadline: Remove needless parameter in dl_runtime_exceeded()
  sched: Remove superfluous resetting of the p->dl_throttled flag
  sched/deadline: Drop duplicate init_sched_dl_class() declaration
  sched/deadline: Reduce rq lock contention by eliminating locking of non-feasible target
  sched/deadline: Make init_sched_dl_class() __init
  sched/deadline: Optimize pull_dl_task()
  sched/preempt: Add static_key() to preempt_notifiers
  sched/preempt: Fix preempt notifiers documentation about hlist_del() within unsafe iteration
  sched/stop_machine: Fix deadlock between multiple stop_two_cpus()
  sched/debug: Add sum_sleep_runtime to /proc/<pid>/sched
  sched/debug: Replace vruntime with wait_sum in /proc/sched_debug
  sched/debug: Properly format runnable tasks in /proc/sched_debug
  sched/numa: Only consider less busy nodes as numa balancing destinations
  Revert 095bebf61a46 ("sched/numa: Do not move past the balance point if unbalanced")
  sched/fair: Prevent throttling in early pick_next_task_fair()
  preempt: Reorganize the notrace definitions a bit
  preempt: Use preempt_schedule_context() as the official tracing preemption point
  sched: Make preempt_schedule_context() function-tracing safe
  x86: Remove cpu_sibling_mask() and cpu_core_mask()
  x86: Replace cpu_**_mask() with topology_**_cpumask()
  ...
---

23b7776290b10297fe2cae0fb5f166a4f2c68121
diff --cc arch/x86/kernel/cpu/perf_event_intel.c
index 19980d9a6cc9,324817735771..b9826a981fb2
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@@ -2594,7 -2639,9 +2594,7 @@@ static void intel_pmu_cpu_starting(int 
  		cpuc->lbr_sel = &cpuc->shared_regs->regs[EXTRA_REG_LBR];
  
  	if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) {
- 		for_each_cpu(i, topology_thread_cpumask(cpu)) {
 -		int h = x86_pmu.num_counters >> 1;
 -
+ 		for_each_cpu(i, topology_sibling_cpumask(cpu)) {
  			struct intel_excl_cntrs *c;
  
  			c = per_cpu(cpu_hw_events, i).excl_cntrs;