[GIT PULL] scheduler changes for v4.7
From: Ingo Molnar
Date: Mon May 16 2016 - 13:08:26 EST
Linus,
Please pull the latest sched-core-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched-core-for-linus
# HEAD: ef0491ea17f8019821c7e9c8e801184ecf17f85a ARM: Hide finish_arch_post_lock_switch() from modules
- massive CPU hotplug rework (Thomas Gleixner)
- improve migration fairness (Peter Zijlstra)
- CPU load calculation updates/cleanups (Yuyang Du)
- cpufreq updates (Steve Muckle)
- nohz optimizations (Frederic Weisbecker)
- switch_mm() micro-optimization on x86 (Andy Lutomirski)
- ... lots of other enhancements, fixes and cleanups.
Thanks,
Ingo
------------------>
Alexander Shishkin (1):
perf/core, sched: Don't use clock function pointer to determine clock
Andy Lutomirski (5):
sched/core, ARM: Include linux/preempt.h from asm/mmu_context.h
sched/core: Add switch_mm_irqs_off() and use it in the scheduler
x86/mm: Build arch/x86/mm/tlb.c even on !SMP
x86/mm, sched/core: Uninline switch_mm()
x86/mm, sched/core: Turn off IRQs in switch_mm()
Anton Blanchard (1):
sched/cpuacct: Check for NULL when using task_pt_regs()
Daniel Lezcano (2):
sched/clock: Remove pointless test in cpu_clock/local_clock
sched/clock: Make local_clock()/cpu_clock() inline
Davidlohr Bueso (1):
sched/core: Fix comment typo in wake_q_add()
Dietmar Eggemann (2):
sched/fair: Remove stale power aware scheduling comments
sched/fair: Fix comment in calculate_imbalance()
Dongsheng Yang (1):
sched/cpuacct: Split usage accounting into user_usage and sys_usage
Frederic Weisbecker (3):
sched/fair: Gather CPU load functions under a more conventional namespace
sched/fair: Correctly handle nohz ticks CPU load accounting
sched/fair: Optimize !CONFIG_NO_HZ_COMMON CPU load updates
Ingo Molnar (1):
mm/mmu_context, sched/core: Fix mmu_context.h assumption
Matt Fleming (1):
sched/fair: Update rq clock before updating nohz CPU load
Morten Rasmussen (1):
sched/fair: Correct unit of load_above_capacity
Muhammad Falak R Wani (1):
sched/core: Remove unused variable
Peter Zijlstra (10):
sched/core: Move task_rq_lock() out of line
sched/core: Introduce 'struct rq_flags'
locking/lockdep, sched/core: Implement a better lock pinning scheme
sched/core: Enable increased load resolution on 64-bit kernels
sched/hotplug: Move sync_rcu to be with set_cpu_active(false)
sched/fair: Move record_wakee()
sched/fair: Prepare to fix fairness problems on migration
sched/core: Kill sched_class::task_waking to clean up the migration logic
sched/fair: Fix fairness issue on migration
sched/fair: Clean up scale confusion
Peter Zijlstra (Intel) (1):
sched: Allow per-cpu kernel threads to run on online && !active
Rabin Vincent (1):
sched/debug: Don't dump sched debug info in SysRq-W
Srikar Dronamraju (2):
sched/fair: Reset nr_balance_failed after active balancing
sched/fair: Fix asym packing to select correct CPU
Steve Muckle (3):
sched/fair: Move cpufreq hook to update_cfs_rq_load_avg()
sched/fair: Do not call cpufreq hook unless util changed
sched/fair: Call cpufreq hook in additional paths
Steven Rostedt (2):
sched/core: Add preempt checks in preempt_schedule() code
ARM: Hide finish_arch_post_lock_switch() from modules
Thomas Gleixner (14):
sched: Make set_cpu_rq_start_time() a built in hotplug state
sched: Allow hotplug notifiers to be setup early
sched: Consolidate the notifier maze
sched: Move sched_domains_numa_masks_clear() to DOWN_PREPARE
sched/hotplug: Convert cpu_[in]active notifiers to state machine
sched/migration: Move prepare transition to SCHED_STARTING state
sched/migration: Move calc_load_migrate() into CPU_DYING
sched/migration: Move CPU_ONLINE into scheduler state
sched/hotplug: Move migration CPU_DYING to sched_cpu_dying()
sched/hotplug: Make activate() the last hotplug step
sched/fair: Make ilb_notifier an explicit call
sched: Make hrtick_notifier an explicit call
sched/core: Use tsk_cpus_allowed() instead of accessing ->cpus_allowed
sched/core: Provide a tsk_nr_cpus_allowed() helper
Tim Chen (1):
sched/numa: Remove unnecessary NUMA dequeue update from non-SMP kernels
Vik Heyndrickx (1):
sched/loadavg: Fix loadavg artifacts on fully idle and on fully loaded systems
Wanpeng Li (3):
sched/cpufreq: Optimize cpufreq update kicker to avoid update multiple times
sched/debug: Print out idle balance values even on !CONFIG_SCHEDSTATS kernels
sched/nohz: Fix affine unpinned timers mess
Xunlei Pang (1):
sched/deadline: Fix a bug in dl_overflow()
Yuyang Du (6):
sched/fair: Update comments after a variable rename
sched/fair: Initiate a new task's util avg to a bounded value
sched/fair: Generalize the load/util averages resolution definition
sched/fair: Rename SCHED_LOAD_SHIFT to NICE_0_LOAD_SHIFT and remove SCHED_LOAD_SCALE
sched/fair: Add detailed description to the sched load avg metrics
sched/fair: Optimize sum computation with a lookup table
Zhao Lei (1):
sched/cpuacct: Show all possible CPUs in cpuacct output
Documentation/trace/ftrace.txt | 10 +-
arch/arm/include/asm/mmu_context.h | 3 +
arch/powerpc/kernel/smp.c | 2 +-
arch/s390/kernel/smp.c | 2 +-
arch/x86/events/core.c | 2 +-
arch/x86/include/asm/mmu_context.h | 101 +----
arch/x86/mm/Makefile | 3 +-
arch/x86/mm/tlb.c | 116 ++++++
include/linux/cpu.h | 18 -
include/linux/cpuhotplug.h | 2 +
include/linux/cpumask.h | 6 +-
include/linux/lockdep.h | 23 +-
include/linux/mmu_context.h | 7 +
include/linux/sched.h | 124 +++++-
kernel/cpu.c | 32 +-
kernel/locking/lockdep.c | 71 +++-
kernel/sched/clock.c | 48 +--
kernel/sched/core.c | 749 +++++++++++++++++++++----------------
kernel/sched/cpuacct.c | 147 ++++++--
kernel/sched/cpudeadline.c | 4 +-
kernel/sched/cpupri.c | 4 +-
kernel/sched/deadline.c | 55 +--
kernel/sched/debug.c | 10 +-
kernel/sched/fair.c | 513 ++++++++++++++++---------
kernel/sched/idle_task.c | 2 +-
kernel/sched/loadavg.c | 11 +-
kernel/sched/rt.c | 38 +-
kernel/sched/sched.h | 140 +++----
kernel/sched/stop_task.c | 2 +-
kernel/time/tick-sched.c | 9 +-
mm/mmu_context.c | 2 +-
31 files changed, 1329 insertions(+), 927 deletions(-)
diff --git a/Documentation/trace/ftrace.txt b/Documentation/trace/ftrace.txt
index f52f297cb406..9857606dd7b7 100644
--- a/Documentation/trace/ftrace.txt
+++ b/Documentation/trace/ftrace.txt
@@ -1562,12 +1562,12 @@ Doing the same with chrt -r 5 and function-trace set.
<idle>-0 3dN.1 12us : menu_hrtimer_cancel <-tick_nohz_idle_exit
<idle>-0 3dN.1 12us : ktime_get <-tick_nohz_idle_exit
<idle>-0 3dN.1 12us : tick_do_update_jiffies64 <-tick_nohz_idle_exit
- <idle>-0 3dN.1 13us : update_cpu_load_nohz <-tick_nohz_idle_exit
- <idle>-0 3dN.1 13us : _raw_spin_lock <-update_cpu_load_nohz
+ <idle>-0 3dN.1 13us : cpu_load_update_nohz <-tick_nohz_idle_exit
+ <idle>-0 3dN.1 13us : _raw_spin_lock <-cpu_load_update_nohz
<idle>-0 3dN.1 13us : add_preempt_count <-_raw_spin_lock
- <idle>-0 3dN.2 13us : __update_cpu_load <-update_cpu_load_nohz
- <idle>-0 3dN.2 14us : sched_avg_update <-__update_cpu_load
- <idle>-0 3dN.2 14us : _raw_spin_unlock <-update_cpu_load_nohz
+ <idle>-0 3dN.2 13us : __cpu_load_update <-cpu_load_update_nohz
+ <idle>-0 3dN.2 14us : sched_avg_update <-__cpu_load_update
+ <idle>-0 3dN.2 14us : _raw_spin_unlock <-cpu_load_update_nohz
<idle>-0 3dN.2 14us : sub_preempt_count <-_raw_spin_unlock
<idle>-0 3dN.1 15us : calc_load_exit_idle <-tick_nohz_idle_exit
<idle>-0 3dN.1 15us : touch_softlockup_watchdog <-tick_nohz_idle_exit
diff --git a/arch/arm/include/asm/mmu_context.h b/arch/arm/include/asm/mmu_context.h
index fa5b42d44985..3cc14dd8587c 100644
--- a/arch/arm/include/asm/mmu_context.h
+++ b/arch/arm/include/asm/mmu_context.h
@@ -15,6 +15,7 @@
#include <linux/compiler.h>
#include <linux/sched.h>
+#include <linux/preempt.h>
#include <asm/cacheflush.h>
#include <asm/cachetype.h>
#include <asm/proc-fns.h>
@@ -66,6 +67,7 @@ static inline void check_and_switch_context(struct mm_struct *mm,
cpu_switch_mm(mm->pgd, mm);
}
+#ifndef MODULE
#define finish_arch_post_lock_switch \
finish_arch_post_lock_switch
static inline void finish_arch_post_lock_switch(void)
@@ -87,6 +89,7 @@ static inline void finish_arch_post_lock_switch(void)
preempt_enable_no_resched();
}
}
+#endif /* !MODULE */
#endif /* CONFIG_MMU */
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 8cac1eb41466..55c924b65f71 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -565,7 +565,7 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle)
smp_ops->give_timebase();
/* Wait until cpu puts itself in the online & active maps */
- while (!cpu_online(cpu) || !cpu_active(cpu))
+ while (!cpu_online(cpu))
cpu_relax();
return 0;
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 40a6b4f9c36c..7b89a7572100 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -832,7 +832,7 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle)
pcpu_attach_task(pcpu, tidle);
pcpu_start_fn(pcpu, smp_start_secondary, NULL);
/* Wait until cpu puts itself in the online & active maps */
- while (!cpu_online(cpu) || !cpu_active(cpu))
+ while (!cpu_online(cpu))
cpu_relax();
return 0;
}
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 041e442a3e28..dd39fde66b54 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2177,7 +2177,7 @@ void arch_perf_update_userpage(struct perf_event *event,
* cap_user_time_zero doesn't make sense when we're using a different
* time base for the records.
*/
- if (event->clock == &local_clock) {
+ if (!event->attr.use_clockid) {
userpg->cap_user_time_zero = 1;
userpg->time_zero = data->cyc2ns_offset;
}
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index 84280029cafd..396348196aa7 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -115,103 +115,12 @@ static inline void destroy_context(struct mm_struct *mm)
destroy_context_ldt(mm);
}
-static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
- struct task_struct *tsk)
-{
- unsigned cpu = smp_processor_id();
+extern void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *tsk);
- if (likely(prev != next)) {
-#ifdef CONFIG_SMP
- this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
- this_cpu_write(cpu_tlbstate.active_mm, next);
-#endif
- cpumask_set_cpu(cpu, mm_cpumask(next));
-
- /*
- * Re-load page tables.
- *
- * This logic has an ordering constraint:
- *
- * CPU 0: Write to a PTE for 'next'
- * CPU 0: load bit 1 in mm_cpumask. if nonzero, send IPI.
- * CPU 1: set bit 1 in next's mm_cpumask
- * CPU 1: load from the PTE that CPU 0 writes (implicit)
- *
- * We need to prevent an outcome in which CPU 1 observes
- * the new PTE value and CPU 0 observes bit 1 clear in
- * mm_cpumask. (If that occurs, then the IPI will never
- * be sent, and CPU 0's TLB will contain a stale entry.)
- *
- * The bad outcome can occur if either CPU's load is
- * reordered before that CPU's store, so both CPUs must
- * execute full barriers to prevent this from happening.
- *
- * Thus, switch_mm needs a full barrier between the
- * store to mm_cpumask and any operation that could load
- * from next->pgd. TLB fills are special and can happen
- * due to instruction fetches or for no reason at all,
- * and neither LOCK nor MFENCE orders them.
- * Fortunately, load_cr3() is serializing and gives the
- * ordering guarantee we need.
- *
- */
- load_cr3(next->pgd);
-
- trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
-
- /* Stop flush ipis for the previous mm */
- cpumask_clear_cpu(cpu, mm_cpumask(prev));
-
- /* Load per-mm CR4 state */
- load_mm_cr4(next);
-
-#ifdef CONFIG_MODIFY_LDT_SYSCALL
- /*
- * Load the LDT, if the LDT is different.
- *
- * It's possible that prev->context.ldt doesn't match
- * the LDT register. This can happen if leave_mm(prev)
- * was called and then modify_ldt changed
- * prev->context.ldt but suppressed an IPI to this CPU.
- * In this case, prev->context.ldt != NULL, because we
- * never set context.ldt to NULL while the mm still
- * exists. That means that next->context.ldt !=
- * prev->context.ldt, because mms never share an LDT.
- */
- if (unlikely(prev->context.ldt != next->context.ldt))
- load_mm_ldt(next);
-#endif
- }
-#ifdef CONFIG_SMP
- else {
- this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
- BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next);
-
- if (!cpumask_test_cpu(cpu, mm_cpumask(next))) {
- /*
- * On established mms, the mm_cpumask is only changed
- * from irq context, from ptep_clear_flush() while in
- * lazy tlb mode, and here. Irqs are blocked during
- * schedule, protecting us from simultaneous changes.
- */
- cpumask_set_cpu(cpu, mm_cpumask(next));
-
- /*
- * We were in lazy tlb mode and leave_mm disabled
- * tlb flush IPI delivery. We must reload CR3
- * to make sure to use no freed page tables.
- *
- * As above, load_cr3() is serializing and orders TLB
- * fills with respect to the mm_cpumask write.
- */
- load_cr3(next->pgd);
- trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
- load_mm_cr4(next);
- load_mm_ldt(next);
- }
- }
-#endif
-}
+extern void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *tsk);
+#define switch_mm_irqs_off switch_mm_irqs_off
#define activate_mm(prev, next) \
do { \
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index f98913258c63..62c0043a5fd5 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -2,7 +2,7 @@
KCOV_INSTRUMENT_tlb.o := n
obj-y := init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
- pat.o pgtable.o physaddr.o gup.o setup_nx.o
+ pat.o pgtable.o physaddr.o gup.o setup_nx.o tlb.o
# Make sure __phys_addr has no stackprotector
nostackp := $(call cc-option, -fno-stack-protector)
@@ -12,7 +12,6 @@ CFLAGS_setup_nx.o := $(nostackp)
CFLAGS_fault.o := -I$(src)/../include/asm/trace
obj-$(CONFIG_X86_PAT) += pat_rbtree.o
-obj-$(CONFIG_SMP) += tlb.o
obj-$(CONFIG_X86_32) += pgtable_32.o iomap_32.o
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index fe9b9f776361..5643fd0b1a7d 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -28,6 +28,8 @@
* Implement flush IPI by CALL_FUNCTION_VECTOR, Alex Shi
*/
+#ifdef CONFIG_SMP
+
struct flush_tlb_info {
struct mm_struct *flush_mm;
unsigned long flush_start;
@@ -57,6 +59,118 @@ void leave_mm(int cpu)
}
EXPORT_SYMBOL_GPL(leave_mm);
+#endif /* CONFIG_SMP */
+
+void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *tsk)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ switch_mm_irqs_off(prev, next, tsk);
+ local_irq_restore(flags);
+}
+
+void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *tsk)
+{
+ unsigned cpu = smp_processor_id();
+
+ if (likely(prev != next)) {
+#ifdef CONFIG_SMP
+ this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
+ this_cpu_write(cpu_tlbstate.active_mm, next);
+#endif
+ cpumask_set_cpu(cpu, mm_cpumask(next));
+
+ /*
+ * Re-load page tables.
+ *
+ * This logic has an ordering constraint:
+ *
+ * CPU 0: Write to a PTE for 'next'
+ * CPU 0: load bit 1 in mm_cpumask. if nonzero, send IPI.
+ * CPU 1: set bit 1 in next's mm_cpumask
+ * CPU 1: load from the PTE that CPU 0 writes (implicit)
+ *
+ * We need to prevent an outcome in which CPU 1 observes
+ * the new PTE value and CPU 0 observes bit 1 clear in
+ * mm_cpumask. (If that occurs, then the IPI will never
+ * be sent, and CPU 0's TLB will contain a stale entry.)
+ *
+ * The bad outcome can occur if either CPU's load is
+ * reordered before that CPU's store, so both CPUs must
+ * execute full barriers to prevent this from happening.
+ *
+ * Thus, switch_mm needs a full barrier between the
+ * store to mm_cpumask and any operation that could load
+ * from next->pgd. TLB fills are special and can happen
+ * due to instruction fetches or for no reason at all,
+ * and neither LOCK nor MFENCE orders them.
+ * Fortunately, load_cr3() is serializing and gives the
+ * ordering guarantee we need.
+ *
+ */
+ load_cr3(next->pgd);
+
+ trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
+
+ /* Stop flush ipis for the previous mm */
+ cpumask_clear_cpu(cpu, mm_cpumask(prev));
+
+ /* Load per-mm CR4 state */
+ load_mm_cr4(next);
+
+#ifdef CONFIG_MODIFY_LDT_SYSCALL
+ /*
+ * Load the LDT, if the LDT is different.
+ *
+ * It's possible that prev->context.ldt doesn't match
+ * the LDT register. This can happen if leave_mm(prev)
+ * was called and then modify_ldt changed
+ * prev->context.ldt but suppressed an IPI to this CPU.
+ * In this case, prev->context.ldt != NULL, because we
+ * never set context.ldt to NULL while the mm still
+ * exists. That means that next->context.ldt !=
+ * prev->context.ldt, because mms never share an LDT.
+ */
+ if (unlikely(prev->context.ldt != next->context.ldt))
+ load_mm_ldt(next);
+#endif
+ }
+#ifdef CONFIG_SMP
+ else {
+ this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
+ BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next);
+
+ if (!cpumask_test_cpu(cpu, mm_cpumask(next))) {
+ /*
+ * On established mms, the mm_cpumask is only changed
+ * from irq context, from ptep_clear_flush() while in
+ * lazy tlb mode, and here. Irqs are blocked during
+ * schedule, protecting us from simultaneous changes.
+ */
+ cpumask_set_cpu(cpu, mm_cpumask(next));
+
+ /*
+ * We were in lazy tlb mode and leave_mm disabled
+ * tlb flush IPI delivery. We must reload CR3
+ * to make sure to use no freed page tables.
+ *
+ * As above, load_cr3() is serializing and orders TLB
+ * fills with respect to the mm_cpumask write.
+ */
+ load_cr3(next->pgd);
+ trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
+ load_mm_cr4(next);
+ load_mm_ldt(next);
+ }
+ }
+#endif
+}
+
+#ifdef CONFIG_SMP
+
/*
* The flush IPI assumes that a thread switch happens in this order:
* [cpu0: the cpu that switches]
@@ -353,3 +467,5 @@ static int __init create_tlb_single_page_flush_ceiling(void)
return 0;
}
late_initcall(create_tlb_single_page_flush_ceiling);
+
+#endif /* CONFIG_SMP */
diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index f9b1fab4388a..21597dcac0e2 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -59,25 +59,7 @@ struct notifier_block;
* CPU notifier priorities.
*/
enum {
- /*
- * SCHED_ACTIVE marks a cpu which is coming up active during
- * CPU_ONLINE and CPU_DOWN_FAILED and must be the first
- * notifier. CPUSET_ACTIVE adjusts cpuset according to
- * cpu_active mask right after SCHED_ACTIVE. During
- * CPU_DOWN_PREPARE, SCHED_INACTIVE and CPUSET_INACTIVE are
- * ordered in the similar way.
- *
- * This ordering guarantees consistent cpu_active mask and
- * migration behavior to all cpu notifiers.
- */
- CPU_PRI_SCHED_ACTIVE = INT_MAX,
- CPU_PRI_CPUSET_ACTIVE = INT_MAX - 1,
- CPU_PRI_SCHED_INACTIVE = INT_MIN + 1,
- CPU_PRI_CPUSET_INACTIVE = INT_MIN,
-
- /* migration should happen before other stuff but after perf */
CPU_PRI_PERF = 20,
- CPU_PRI_MIGRATION = 10,
/* bring up workqueues before normal notifiers and down after */
CPU_PRI_WORKQUEUE_UP = 5,
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 5d68e15e46b7..386374d19987 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -8,6 +8,7 @@ enum cpuhp_state {
CPUHP_BRINGUP_CPU,
CPUHP_AP_IDLE_DEAD,
CPUHP_AP_OFFLINE,
+ CPUHP_AP_SCHED_STARTING,
CPUHP_AP_NOTIFY_STARTING,
CPUHP_AP_ONLINE,
CPUHP_TEARDOWN_CPU,
@@ -16,6 +17,7 @@ enum cpuhp_state {
CPUHP_AP_NOTIFY_ONLINE,
CPUHP_AP_ONLINE_DYN,
CPUHP_AP_ONLINE_DYN_END = CPUHP_AP_ONLINE_DYN + 30,
+ CPUHP_AP_ACTIVE,
CPUHP_ONLINE,
};
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 40cee6b77a93..e828cf65d7df 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -743,12 +743,10 @@ set_cpu_present(unsigned int cpu, bool present)
static inline void
set_cpu_online(unsigned int cpu, bool online)
{
- if (online) {
+ if (online)
cpumask_set_cpu(cpu, &__cpu_online_mask);
- cpumask_set_cpu(cpu, &__cpu_active_mask);
- } else {
+ else
cpumask_clear_cpu(cpu, &__cpu_online_mask);
- }
}
static inline void
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index d10ef06971b5..fb7d87e45fbe 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -356,8 +356,13 @@ extern void lockdep_set_current_reclaim_state(gfp_t gfp_mask);
extern void lockdep_clear_current_reclaim_state(void);
extern void lockdep_trace_alloc(gfp_t mask);
-extern void lock_pin_lock(struct lockdep_map *lock);
-extern void lock_unpin_lock(struct lockdep_map *lock);
+struct pin_cookie { unsigned int val; };
+
+#define NIL_COOKIE (struct pin_cookie){ .val = 0U, }
+
+extern struct pin_cookie lock_pin_lock(struct lockdep_map *lock);
+extern void lock_repin_lock(struct lockdep_map *lock, struct pin_cookie);
+extern void lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie);
# define INIT_LOCKDEP .lockdep_recursion = 0, .lockdep_reclaim_gfp = 0,
@@ -373,8 +378,9 @@ extern void lock_unpin_lock(struct lockdep_map *lock);
#define lockdep_recursing(tsk) ((tsk)->lockdep_recursion)
-#define lockdep_pin_lock(l) lock_pin_lock(&(l)->dep_map)
-#define lockdep_unpin_lock(l) lock_unpin_lock(&(l)->dep_map)
+#define lockdep_pin_lock(l) lock_pin_lock(&(l)->dep_map)
+#define lockdep_repin_lock(l,c) lock_repin_lock(&(l)->dep_map, (c))
+#define lockdep_unpin_lock(l,c) lock_unpin_lock(&(l)->dep_map, (c))
#else /* !CONFIG_LOCKDEP */
@@ -427,8 +433,13 @@ struct lock_class_key { };
#define lockdep_recursing(tsk) (0)
-#define lockdep_pin_lock(l) do { (void)(l); } while (0)
-#define lockdep_unpin_lock(l) do { (void)(l); } while (0)
+struct pin_cookie { };
+
+#define NIL_COOKIE (struct pin_cookie){ }
+
+#define lockdep_pin_lock(l) ({ struct pin_cookie cookie; cookie; })
+#define lockdep_repin_lock(l, c) do { (void)(l); (void)(c); } while (0)
+#define lockdep_unpin_lock(l, c) do { (void)(l); (void)(c); } while (0)
#endif /* !LOCKDEP */
diff --git a/include/linux/mmu_context.h b/include/linux/mmu_context.h
index 70fffeba7495..a4441784503b 100644
--- a/include/linux/mmu_context.h
+++ b/include/linux/mmu_context.h
@@ -1,9 +1,16 @@
#ifndef _LINUX_MMU_CONTEXT_H
#define _LINUX_MMU_CONTEXT_H
+#include <asm/mmu_context.h>
+
struct mm_struct;
void use_mm(struct mm_struct *mm);
void unuse_mm(struct mm_struct *mm);
+/* Architectures that care about IRQ state in switch_mm can override this. */
+#ifndef switch_mm_irqs_off
+# define switch_mm_irqs_off switch_mm
+#endif
+
#endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 52c4847b05e2..38526b67e787 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -178,9 +178,11 @@ extern void get_iowait_load(unsigned long *nr_waiters, unsigned long *load);
extern void calc_global_load(unsigned long ticks);
#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
-extern void update_cpu_load_nohz(int active);
+extern void cpu_load_update_nohz_start(void);
+extern void cpu_load_update_nohz_stop(void);
#else
-static inline void update_cpu_load_nohz(int active) { }
+static inline void cpu_load_update_nohz_start(void) { }
+static inline void cpu_load_update_nohz_stop(void) { }
#endif
extern void dump_cpu_task(int cpu);
@@ -372,6 +374,15 @@ extern void cpu_init (void);
extern void trap_init(void);
extern void update_process_times(int user);
extern void scheduler_tick(void);
+extern int sched_cpu_starting(unsigned int cpu);
+extern int sched_cpu_activate(unsigned int cpu);
+extern int sched_cpu_deactivate(unsigned int cpu);
+
+#ifdef CONFIG_HOTPLUG_CPU
+extern int sched_cpu_dying(unsigned int cpu);
+#else
+# define sched_cpu_dying NULL
+#endif
extern void sched_show_task(struct task_struct *p);
@@ -935,9 +946,19 @@ enum cpu_idle_type {
};
/*
+ * Integer metrics need fixed point arithmetic, e.g., sched/fair
+ * has a few: load, load_avg, util_avg, freq, and capacity.
+ *
+ * We define a basic fixed point arithmetic range, and then formalize
+ * all these metrics based on that basic range.
+ */
+# define SCHED_FIXEDPOINT_SHIFT 10
+# define SCHED_FIXEDPOINT_SCALE (1L << SCHED_FIXEDPOINT_SHIFT)
+
+/*
* Increase resolution of cpu_capacity calculations
*/
-#define SCHED_CAPACITY_SHIFT 10
+#define SCHED_CAPACITY_SHIFT SCHED_FIXEDPOINT_SHIFT
#define SCHED_CAPACITY_SCALE (1L << SCHED_CAPACITY_SHIFT)
/*
@@ -1199,18 +1220,56 @@ struct load_weight {
};
/*
- * The load_avg/util_avg accumulates an infinite geometric series.
- * 1) load_avg factors frequency scaling into the amount of time that a
- * sched_entity is runnable on a rq into its weight. For cfs_rq, it is the
- * aggregated such weights of all runnable and blocked sched_entities.
- * 2) util_avg factors frequency and cpu scaling into the amount of time
- * that a sched_entity is running on a CPU, in the range [0..SCHED_LOAD_SCALE].
- * For cfs_rq, it is the aggregated such times of all runnable and
+ * The load_avg/util_avg accumulates an infinite geometric series
+ * (see __update_load_avg() in kernel/sched/fair.c).
+ *
+ * [load_avg definition]
+ *
+ * load_avg = runnable% * scale_load_down(load)
+ *
+ * where runnable% is the time ratio that a sched_entity is runnable.
+ * For cfs_rq, it is the aggregated load_avg of all runnable and
* blocked sched_entities.
- * The 64 bit load_sum can:
- * 1) for cfs_rq, afford 4353082796 (=2^64/47742/88761) entities with
- * the highest weight (=88761) always runnable, we should not overflow
- * 2) for entity, support any load.weight always runnable
+ *
+ * load_avg may also take frequency scaling into account:
+ *
+ * load_avg = runnable% * scale_load_down(load) * freq%
+ *
+ * where freq% is the CPU frequency normalized to the highest frequency.
+ *
+ * [util_avg definition]
+ *
+ * util_avg = running% * SCHED_CAPACITY_SCALE
+ *
+ * where running% is the time ratio that a sched_entity is running on
+ * a CPU. For cfs_rq, it is the aggregated util_avg of all runnable
+ * and blocked sched_entities.
+ *
+ * util_avg may also factor frequency scaling and CPU capacity scaling:
+ *
+ * util_avg = running% * SCHED_CAPACITY_SCALE * freq% * capacity%
+ *
+ * where freq% is the same as above, and capacity% is the CPU capacity
+ * normalized to the greatest capacity (due to uarch differences, etc).
+ *
+ * N.B., the above ratios (runnable%, running%, freq%, and capacity%)
+ * themselves are in the range of [0, 1]. To do fixed point arithmetics,
+ * we therefore scale them to as large a range as necessary. This is for
+ * example reflected by util_avg's SCHED_CAPACITY_SCALE.
+ *
+ * [Overflow issue]
+ *
+ * The 64-bit load_sum can have 4353082796 (=2^64/47742/88761) entities
+ * with the highest load (=88761), always runnable on a single cfs_rq,
+ * and should not overflow as the number already hits PID_MAX_LIMIT.
+ *
+ * For all other cases (including 32-bit kernels), struct load_weight's
+ * weight will overflow first before we do, because:
+ *
+ * Max(load_avg) <= Max(load.weight)
+ *
+ * Then it is the load_weight's responsibility to consider overflow
+ * issues.
*/
struct sched_avg {
u64 last_update_time, load_sum;
@@ -1871,6 +1930,11 @@ extern int arch_task_struct_size __read_mostly;
/* Future-safe accessor for struct task_struct's cpus_allowed. */
#define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed)
+static inline int tsk_nr_cpus_allowed(struct task_struct *p)
+{
+ return p->nr_cpus_allowed;
+}
+
#define TNF_MIGRATED 0x01
#define TNF_NO_GROUP 0x02
#define TNF_SHARED 0x04
@@ -2303,8 +2367,6 @@ extern unsigned long long notrace sched_clock(void);
/*
* See the comment in kernel/sched/clock.c
*/
-extern u64 cpu_clock(int cpu);
-extern u64 local_clock(void);
extern u64 running_clock(void);
extern u64 sched_clock_cpu(int cpu);
@@ -2323,6 +2385,16 @@ static inline void sched_clock_idle_sleep_event(void)
static inline void sched_clock_idle_wakeup_event(u64 delta_ns)
{
}
+
+static inline u64 cpu_clock(int cpu)
+{
+ return sched_clock();
+}
+
+static inline u64 local_clock(void)
+{
+ return sched_clock();
+}
#else
/*
* Architectures can set this to 1 if they have specified
@@ -2337,6 +2409,26 @@ extern void clear_sched_clock_stable(void);
extern void sched_clock_tick(void);
extern void sched_clock_idle_sleep_event(void);
extern void sched_clock_idle_wakeup_event(u64 delta_ns);
+
+/*
+ * As outlined in clock.c, provides a fast, high resolution, nanosecond
+ * time source that is monotonic per cpu argument and has bounded drift
+ * between cpus.
+ *
+ * ######################### BIG FAT WARNING ##########################
+ * # when comparing cpu_clock(i) to cpu_clock(j) for i != j, time can #
+ * # go backwards !! #
+ * ####################################################################
+ */
+static inline u64 cpu_clock(int cpu)
+{
+ return sched_clock_cpu(cpu);
+}
+
+static inline u64 local_clock(void)
+{
+ return sched_clock_cpu(raw_smp_processor_id());
+}
#endif
#ifdef CONFIG_IRQ_TIME_ACCOUNTING
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 3e3f6e49eabb..d948e44c471e 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -703,21 +703,6 @@ static int takedown_cpu(unsigned int cpu)
struct cpuhp_cpu_state *st = per_cpu_ptr(&cpuhp_state, cpu);
int err;
- /*
- * By now we've cleared cpu_active_mask, wait for all preempt-disabled
- * and RCU users of this state to go away such that all new such users
- * will observe it.
- *
- * For CONFIG_PREEMPT we have preemptible RCU and its sync_rcu() might
- * not imply sync_sched(), so wait for both.
- *
- * Do sync before park smpboot threads to take care the rcu boost case.
- */
- if (IS_ENABLED(CONFIG_PREEMPT))
- synchronize_rcu_mult(call_rcu, call_rcu_sched);
- else
- synchronize_rcu();
-
/* Park the smpboot threads */
kthread_park(per_cpu_ptr(&cpuhp_state, cpu)->thread);
smpboot_park_threads(cpu);
@@ -923,8 +908,6 @@ void cpuhp_online_idle(enum cpuhp_state state)
st->state = CPUHP_AP_ONLINE_IDLE;
- /* The cpu is marked online, set it active now */
- set_cpu_active(cpu, true);
/* Unpark the stopper thread and the hotplug thread of this cpu */
stop_machine_unpark(cpu);
kthread_unpark(st->thread);
@@ -1236,6 +1219,12 @@ static struct cpuhp_step cpuhp_ap_states[] = {
.name = "ap:offline",
.cant_stop = true,
},
+ /* First state is scheduler control. Interrupts are disabled */
+ [CPUHP_AP_SCHED_STARTING] = {
+ .name = "sched:starting",
+ .startup = sched_cpu_starting,
+ .teardown = sched_cpu_dying,
+ },
/*
* Low level startup/teardown notifiers. Run with interrupts
* disabled. Will be removed once the notifiers are converted to
@@ -1274,6 +1263,15 @@ static struct cpuhp_step cpuhp_ap_states[] = {
* The dynamically registered state space is here
*/
+#ifdef CONFIG_SMP
+ /* Last state is scheduler control setting the cpu active */
+ [CPUHP_AP_ACTIVE] = {
+ .name = "sched:active",
+ .startup = sched_cpu_activate,
+ .teardown = sched_cpu_deactivate,
+ },
+#endif
+
/* CPU is fully up and running. */
[CPUHP_ONLINE] = {
.name = "online",
diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 78c1c0ee6dc1..68bc6a654ca3 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -45,6 +45,7 @@
#include <linux/bitops.h>
#include <linux/gfp.h>
#include <linux/kmemcheck.h>
+#include <linux/random.h>
#include <asm/sections.h>
@@ -3585,7 +3586,35 @@ static int __lock_is_held(struct lockdep_map *lock)
return 0;
}
-static void __lock_pin_lock(struct lockdep_map *lock)
+static struct pin_cookie __lock_pin_lock(struct lockdep_map *lock)
+{
+ struct pin_cookie cookie = NIL_COOKIE;
+ struct task_struct *curr = current;
+ int i;
+
+ if (unlikely(!debug_locks))
+ return cookie;
+
+ for (i = 0; i < curr->lockdep_depth; i++) {
+ struct held_lock *hlock = curr->held_locks + i;
+
+ if (match_held_lock(hlock, lock)) {
+ /*
+ * Grab 16bits of randomness; this is sufficient to not
+ * be guessable and still allows some pin nesting in
+ * our u32 pin_count.
+ */
+ cookie.val = 1 + (prandom_u32() >> 16);
+ hlock->pin_count += cookie.val;
+ return cookie;
+ }
+ }
+
+ WARN(1, "pinning an unheld lock\n");
+ return cookie;
+}
+
+static void __lock_repin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
{
struct task_struct *curr = current;
int i;
@@ -3597,7 +3626,7 @@ static void __lock_pin_lock(struct lockdep_map *lock)
struct held_lock *hlock = curr->held_locks + i;
if (match_held_lock(hlock, lock)) {
- hlock->pin_count++;
+ hlock->pin_count += cookie.val;
return;
}
}
@@ -3605,7 +3634,7 @@ static void __lock_pin_lock(struct lockdep_map *lock)
WARN(1, "pinning an unheld lock\n");
}
-static void __lock_unpin_lock(struct lockdep_map *lock)
+static void __lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
{
struct task_struct *curr = current;
int i;
@@ -3620,7 +3649,11 @@ static void __lock_unpin_lock(struct lockdep_map *lock)
if (WARN(!hlock->pin_count, "unpinning an unpinned lock\n"))
return;
- hlock->pin_count--;
+ hlock->pin_count -= cookie.val;
+
+ if (WARN((int)hlock->pin_count < 0, "pin count corrupted\n"))
+ hlock->pin_count = 0;
+
return;
}
}
@@ -3751,24 +3784,44 @@ int lock_is_held(struct lockdep_map *lock)
}
EXPORT_SYMBOL_GPL(lock_is_held);
-void lock_pin_lock(struct lockdep_map *lock)
+struct pin_cookie lock_pin_lock(struct lockdep_map *lock)
{
+ struct pin_cookie cookie = NIL_COOKIE;
unsigned long flags;
if (unlikely(current->lockdep_recursion))
- return;
+ return cookie;
raw_local_irq_save(flags);
check_flags(flags);
current->lockdep_recursion = 1;
- __lock_pin_lock(lock);
+ cookie = __lock_pin_lock(lock);
current->lockdep_recursion = 0;
raw_local_irq_restore(flags);
+
+ return cookie;
}
EXPORT_SYMBOL_GPL(lock_pin_lock);
-void lock_unpin_lock(struct lockdep_map *lock)
+void lock_repin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
+{
+ unsigned long flags;
+
+ if (unlikely(current->lockdep_recursion))
+ return;
+
+ raw_local_irq_save(flags);
+ check_flags(flags);
+
+ current->lockdep_recursion = 1;
+ __lock_repin_lock(lock, cookie);
+ current->lockdep_recursion = 0;
+ raw_local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(lock_repin_lock);
+
+void lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
{
unsigned long flags;
@@ -3779,7 +3832,7 @@ void lock_unpin_lock(struct lockdep_map *lock)
check_flags(flags);
current->lockdep_recursion = 1;
- __lock_unpin_lock(lock);
+ __lock_unpin_lock(lock, cookie);
current->lockdep_recursion = 0;
raw_local_irq_restore(flags);
}
diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c
index fedb967a9841..e85a725e5c34 100644
--- a/kernel/sched/clock.c
+++ b/kernel/sched/clock.c
@@ -318,6 +318,7 @@ u64 sched_clock_cpu(int cpu)
return clock;
}
+EXPORT_SYMBOL_GPL(sched_clock_cpu);
void sched_clock_tick(void)
{
@@ -363,39 +364,6 @@ void sched_clock_idle_wakeup_event(u64 delta_ns)
}
EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
-/*
- * As outlined at the top, provides a fast, high resolution, nanosecond
- * time source that is monotonic per cpu argument and has bounded drift
- * between cpus.
- *
- * ######################### BIG FAT WARNING ##########################
- * # when comparing cpu_clock(i) to cpu_clock(j) for i != j, time can #
- * # go backwards !! #
- * ####################################################################
- */
-u64 cpu_clock(int cpu)
-{
- if (!sched_clock_stable())
- return sched_clock_cpu(cpu);
-
- return sched_clock();
-}
-
-/*
- * Similar to cpu_clock() for the current cpu. Time will only be observed
- * to be monotonic if care is taken to only compare timestampt taken on the
- * same CPU.
- *
- * See cpu_clock().
- */
-u64 local_clock(void)
-{
- if (!sched_clock_stable())
- return sched_clock_cpu(raw_smp_processor_id());
-
- return sched_clock();
-}
-
#else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
void sched_clock_init(void)
@@ -410,22 +378,8 @@ u64 sched_clock_cpu(int cpu)
return sched_clock();
}
-
-u64 cpu_clock(int cpu)
-{
- return sched_clock();
-}
-
-u64 local_clock(void)
-{
- return sched_clock();
-}
-
#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
-EXPORT_SYMBOL_GPL(cpu_clock);
-EXPORT_SYMBOL_GPL(local_clock);
-
/*
* Running clock - returns the time that has elapsed while a guest has been
* running.
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d1f7149f8704..404c0784b1fc 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -33,7 +33,7 @@
#include <linux/init.h>
#include <linux/uaccess.h>
#include <linux/highmem.h>
-#include <asm/mmu_context.h>
+#include <linux/mmu_context.h>
#include <linux/interrupt.h>
#include <linux/capability.h>
#include <linux/completion.h>
@@ -170,6 +170,71 @@ static struct rq *this_rq_lock(void)
return rq;
}
+/*
+ * __task_rq_lock - lock the rq @p resides on.
+ */
+struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf)
+ __acquires(rq->lock)
+{
+ struct rq *rq;
+
+ lockdep_assert_held(&p->pi_lock);
+
+ for (;;) {
+ rq = task_rq(p);
+ raw_spin_lock(&rq->lock);
+ if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) {
+ rf->cookie = lockdep_pin_lock(&rq->lock);
+ return rq;
+ }
+ raw_spin_unlock(&rq->lock);
+
+ while (unlikely(task_on_rq_migrating(p)))
+ cpu_relax();
+ }
+}
+
+/*
+ * task_rq_lock - lock p->pi_lock and lock the rq @p resides on.
+ */
+struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf)
+ __acquires(p->pi_lock)
+ __acquires(rq->lock)
+{
+ struct rq *rq;
+
+ for (;;) {
+ raw_spin_lock_irqsave(&p->pi_lock, rf->flags);
+ rq = task_rq(p);
+ raw_spin_lock(&rq->lock);
+ /*
+ * move_queued_task() task_rq_lock()
+ *
+ * ACQUIRE (rq->lock)
+ * [S] ->on_rq = MIGRATING [L] rq = task_rq()
+ * WMB (__set_task_cpu()) ACQUIRE (rq->lock);
+ * [S] ->cpu = new_cpu [L] task_rq()
+ * [L] ->on_rq
+ * RELEASE (rq->lock)
+ *
+ * If we observe the old cpu in task_rq_lock, the acquire of
+ * the old rq->lock will fully serialize against the stores.
+ *
+ * If we observe the new cpu in task_rq_lock, the acquire will
+ * pair with the WMB to ensure we must then also see migrating.
+ */
+ if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) {
+ rf->cookie = lockdep_pin_lock(&rq->lock);
+ return rq;
+ }
+ raw_spin_unlock(&rq->lock);
+ raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags);
+
+ while (unlikely(task_on_rq_migrating(p)))
+ cpu_relax();
+ }
+}
+
#ifdef CONFIG_SCHED_HRTICK
/*
* Use HR-timers to deliver accurate preemption points.
@@ -249,29 +314,6 @@ void hrtick_start(struct rq *rq, u64 delay)
}
}
-static int
-hotplug_hrtick(struct notifier_block *nfb, unsigned long action, void *hcpu)
-{
- int cpu = (int)(long)hcpu;
-
- switch (action) {
- case CPU_UP_CANCELED:
- case CPU_UP_CANCELED_FROZEN:
- case CPU_DOWN_PREPARE:
- case CPU_DOWN_PREPARE_FROZEN:
- case CPU_DEAD:
- case CPU_DEAD_FROZEN:
- hrtick_clear(cpu_rq(cpu));
- return NOTIFY_OK;
- }
-
- return NOTIFY_DONE;
-}
-
-static __init void init_hrtick(void)
-{
- hotcpu_notifier(hotplug_hrtick, 0);
-}
#else
/*
* Called to set the hrtick timer state.
@@ -288,10 +330,6 @@ void hrtick_start(struct rq *rq, u64 delay)
hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay),
HRTIMER_MODE_REL_PINNED);
}
-
-static inline void init_hrtick(void)
-{
-}
#endif /* CONFIG_SMP */
static void init_rq_hrtick(struct rq *rq)
@@ -315,10 +353,6 @@ static inline void hrtick_clear(struct rq *rq)
static inline void init_rq_hrtick(struct rq *rq)
{
}
-
-static inline void init_hrtick(void)
-{
-}
#endif /* CONFIG_SCHED_HRTICK */
/*
@@ -400,7 +434,7 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
* wakeup due to that.
*
* This cmpxchg() implies a full barrier, which pairs with the write
- * barrier implied by the wakeup in wake_up_list().
+ * barrier implied by the wakeup in wake_up_q().
*/
if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL))
return;
@@ -499,7 +533,10 @@ int get_nohz_timer_target(void)
rcu_read_lock();
for_each_domain(cpu, sd) {
for_each_cpu(i, sched_domain_span(sd)) {
- if (!idle_cpu(i) && is_housekeeping_cpu(cpu)) {
+ if (cpu == i)
+ continue;
+
+ if (!idle_cpu(i) && is_housekeeping_cpu(i)) {
cpu = i;
goto unlock;
}
@@ -1085,12 +1122,20 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask)
static int __set_cpus_allowed_ptr(struct task_struct *p,
const struct cpumask *new_mask, bool check)
{
- unsigned long flags;
- struct rq *rq;
+ const struct cpumask *cpu_valid_mask = cpu_active_mask;
unsigned int dest_cpu;
+ struct rq_flags rf;
+ struct rq *rq;
int ret = 0;
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
+
+ if (p->flags & PF_KTHREAD) {
+ /*
+ * Kernel threads are allowed on online && !active CPUs
+ */
+ cpu_valid_mask = cpu_online_mask;
+ }
/*
* Must re-check here, to close a race against __kthread_bind(),
@@ -1104,22 +1149,32 @@ static int __set_cpus_allowed_ptr(struct task_struct *p,
if (cpumask_equal(&p->cpus_allowed, new_mask))
goto out;
- if (!cpumask_intersects(new_mask, cpu_active_mask)) {
+ if (!cpumask_intersects(new_mask, cpu_valid_mask)) {
ret = -EINVAL;
goto out;
}
do_set_cpus_allowed(p, new_mask);
+ if (p->flags & PF_KTHREAD) {
+ /*
+ * For kernel threads that do indeed end up on online &&
+ * !active we want to ensure they are strict per-cpu threads.
+ */
+ WARN_ON(cpumask_intersects(new_mask, cpu_online_mask) &&
+ !cpumask_intersects(new_mask, cpu_active_mask) &&
+ p->nr_cpus_allowed != 1);
+ }
+
/* Can the task run on the task's current CPU? If so, we're done */
if (cpumask_test_cpu(task_cpu(p), new_mask))
goto out;
- dest_cpu = cpumask_any_and(cpu_active_mask, new_mask);
+ dest_cpu = cpumask_any_and(cpu_valid_mask, new_mask);
if (task_running(rq, p) || p->state == TASK_WAKING) {
struct migration_arg arg = { p, dest_cpu };
/* Need help from migration thread: drop lock and wait. */
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg);
tlb_migrate_finish(p->mm);
return 0;
@@ -1128,12 +1183,12 @@ static int __set_cpus_allowed_ptr(struct task_struct *p,
* OK, since we're going to drop the lock immediately
* afterwards anyway.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, rf.cookie);
rq = move_queued_task(rq, p, dest_cpu);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, rf.cookie);
}
out:
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return ret;
}
@@ -1317,8 +1372,8 @@ int migrate_swap(struct task_struct *cur, struct task_struct *p)
*/
unsigned long wait_task_inactive(struct task_struct *p, long match_state)
{
- unsigned long flags;
int running, queued;
+ struct rq_flags rf;
unsigned long ncsw;
struct rq *rq;
@@ -1353,14 +1408,14 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
* lock now, to be *sure*. If we're wrong, we'll
* just go back and repeat.
*/
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
trace_sched_wait_task(p);
running = task_running(rq, p);
queued = task_on_rq_queued(p);
ncsw = 0;
if (!match_state || p->state == match_state)
ncsw = p->nvcsw | LONG_MIN; /* sets MSB */
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
/*
* If it changed from the expected state, bail out now.
@@ -1434,6 +1489,25 @@ EXPORT_SYMBOL_GPL(kick_process);
/*
* ->cpus_allowed is protected by both rq->lock and p->pi_lock
+ *
+ * A few notes on cpu_active vs cpu_online:
+ *
+ * - cpu_active must be a subset of cpu_online
+ *
+ * - on cpu-up we allow per-cpu kthreads on the online && !active cpu,
+ * see __set_cpus_allowed_ptr(). At this point the newly online
+ * cpu isn't yet part of the sched domains, and balancing will not
+ * see it.
+ *
+ * - on cpu-down we clear cpu_active() to mask the sched domains and
+ * avoid the load balancer to place new tasks on the to be removed
+ * cpu. Existing tasks will remain running there and will be taken
+ * off.
+ *
+ * This means that fallback selection must not select !active CPUs.
+ * And can assume that any active CPU must be online. Conversely
+ * select_task_rq() below may allow selection of !active CPUs in order
+ * to satisfy the above rules.
*/
static int select_fallback_rq(int cpu, struct task_struct *p)
{
@@ -1452,8 +1526,6 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
/* Look for allowed, online CPU in same node. */
for_each_cpu(dest_cpu, nodemask) {
- if (!cpu_online(dest_cpu))
- continue;
if (!cpu_active(dest_cpu))
continue;
if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p)))
@@ -1464,8 +1536,6 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
for (;;) {
/* Any allowed, online CPU? */
for_each_cpu(dest_cpu, tsk_cpus_allowed(p)) {
- if (!cpu_online(dest_cpu))
- continue;
if (!cpu_active(dest_cpu))
continue;
goto out;
@@ -1515,8 +1585,10 @@ int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags)
{
lockdep_assert_held(&p->pi_lock);
- if (p->nr_cpus_allowed > 1)
+ if (tsk_nr_cpus_allowed(p) > 1)
cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags);
+ else
+ cpu = cpumask_any(tsk_cpus_allowed(p));
/*
* In order not to call set_task_cpu() on a blocking task we need
@@ -1604,8 +1676,8 @@ static inline void ttwu_activate(struct rq *rq, struct task_struct *p, int en_fl
/*
* Mark the task runnable and perform wakeup-preemption.
*/
-static void
-ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags)
+static void ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags,
+ struct pin_cookie cookie)
{
check_preempt_curr(rq, p, wake_flags);
p->state = TASK_RUNNING;
@@ -1617,9 +1689,9 @@ ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags)
* Our task @p is fully woken up and running; so its safe to
* drop the rq->lock, hereafter rq is only used for statistics.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
p->sched_class->task_woken(rq, p);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, cookie);
}
if (rq->idle_stamp) {
@@ -1637,17 +1709,23 @@ ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags)
}
static void
-ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags)
+ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags,
+ struct pin_cookie cookie)
{
+ int en_flags = ENQUEUE_WAKEUP;
+
lockdep_assert_held(&rq->lock);
#ifdef CONFIG_SMP
if (p->sched_contributes_to_load)
rq->nr_uninterruptible--;
+
+ if (wake_flags & WF_MIGRATED)
+ en_flags |= ENQUEUE_MIGRATED;
#endif
- ttwu_activate(rq, p, ENQUEUE_WAKEUP | ENQUEUE_WAKING);
- ttwu_do_wakeup(rq, p, wake_flags);
+ ttwu_activate(rq, p, en_flags);
+ ttwu_do_wakeup(rq, p, wake_flags, cookie);
}
/*
@@ -1658,17 +1736,18 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags)
*/
static int ttwu_remote(struct task_struct *p, int wake_flags)
{
+ struct rq_flags rf;
struct rq *rq;
int ret = 0;
- rq = __task_rq_lock(p);
+ rq = __task_rq_lock(p, &rf);
if (task_on_rq_queued(p)) {
/* check_preempt_curr() may use rq clock */
update_rq_clock(rq);
- ttwu_do_wakeup(rq, p, wake_flags);
+ ttwu_do_wakeup(rq, p, wake_flags, rf.cookie);
ret = 1;
}
- __task_rq_unlock(rq);
+ __task_rq_unlock(rq, &rf);
return ret;
}
@@ -1678,6 +1757,7 @@ void sched_ttwu_pending(void)
{
struct rq *rq = this_rq();
struct llist_node *llist = llist_del_all(&rq->wake_list);
+ struct pin_cookie cookie;
struct task_struct *p;
unsigned long flags;
@@ -1685,15 +1765,19 @@ void sched_ttwu_pending(void)
return;
raw_spin_lock_irqsave(&rq->lock, flags);
- lockdep_pin_lock(&rq->lock);
+ cookie = lockdep_pin_lock(&rq->lock);
while (llist) {
p = llist_entry(llist, struct task_struct, wake_entry);
llist = llist_next(llist);
- ttwu_do_activate(rq, p, 0);
+ /*
+ * See ttwu_queue(); we only call ttwu_queue_remote() when
+ * its a x-cpu wakeup.
+ */
+ ttwu_do_activate(rq, p, WF_MIGRATED, cookie);
}
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
raw_spin_unlock_irqrestore(&rq->lock, flags);
}
@@ -1777,9 +1861,10 @@ bool cpus_share_cache(int this_cpu, int that_cpu)
}
#endif /* CONFIG_SMP */
-static void ttwu_queue(struct task_struct *p, int cpu)
+static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags)
{
struct rq *rq = cpu_rq(cpu);
+ struct pin_cookie cookie;
#if defined(CONFIG_SMP)
if (sched_feat(TTWU_QUEUE) && !cpus_share_cache(smp_processor_id(), cpu)) {
@@ -1790,9 +1875,9 @@ static void ttwu_queue(struct task_struct *p, int cpu)
#endif
raw_spin_lock(&rq->lock);
- lockdep_pin_lock(&rq->lock);
- ttwu_do_activate(rq, p, 0);
- lockdep_unpin_lock(&rq->lock);
+ cookie = lockdep_pin_lock(&rq->lock);
+ ttwu_do_activate(rq, p, wake_flags, cookie);
+ lockdep_unpin_lock(&rq->lock, cookie);
raw_spin_unlock(&rq->lock);
}
@@ -1961,9 +2046,6 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
p->sched_contributes_to_load = !!task_contributes_to_load(p);
p->state = TASK_WAKING;
- if (p->sched_class->task_waking)
- p->sched_class->task_waking(p);
-
cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags);
if (task_cpu(p) != cpu) {
wake_flags |= WF_MIGRATED;
@@ -1971,7 +2053,7 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
}
#endif /* CONFIG_SMP */
- ttwu_queue(p, cpu);
+ ttwu_queue(p, cpu, wake_flags);
stat:
if (schedstat_enabled())
ttwu_stat(p, cpu, wake_flags);
@@ -1989,7 +2071,7 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
* ensure that this_rq() is locked, @p is bound to this_rq() and not
* the current task.
*/
-static void try_to_wake_up_local(struct task_struct *p)
+static void try_to_wake_up_local(struct task_struct *p, struct pin_cookie cookie)
{
struct rq *rq = task_rq(p);
@@ -2006,11 +2088,11 @@ static void try_to_wake_up_local(struct task_struct *p)
* disabled avoiding further scheduler activity on it and we've
* not yet picked a replacement task.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
raw_spin_unlock(&rq->lock);
raw_spin_lock(&p->pi_lock);
raw_spin_lock(&rq->lock);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, cookie);
}
if (!(p->state & TASK_NORMAL))
@@ -2021,7 +2103,7 @@ static void try_to_wake_up_local(struct task_struct *p)
if (!task_on_rq_queued(p))
ttwu_activate(rq, p, ENQUEUE_WAKEUP);
- ttwu_do_wakeup(rq, p, 0);
+ ttwu_do_wakeup(rq, p, 0, cookie);
if (schedstat_enabled())
ttwu_stat(p, smp_processor_id(), 0);
out:
@@ -2381,7 +2463,8 @@ static int dl_overflow(struct task_struct *p, int policy,
u64 new_bw = dl_policy(policy) ? to_ratio(period, runtime) : 0;
int cpus, err = -1;
- if (new_bw == p->dl.dl_bw)
+ /* !deadline task may carry old deadline bandwidth */
+ if (new_bw == p->dl.dl_bw && task_has_dl_policy(p))
return 0;
/*
@@ -2420,12 +2503,12 @@ extern void init_dl_bw(struct dl_bw *dl_b);
*/
void wake_up_new_task(struct task_struct *p)
{
- unsigned long flags;
+ struct rq_flags rf;
struct rq *rq;
- raw_spin_lock_irqsave(&p->pi_lock, flags);
/* Initialize new task's runnable average */
init_entity_runnable_average(&p->se);
+ raw_spin_lock_irqsave(&p->pi_lock, rf.flags);
#ifdef CONFIG_SMP
/*
* Fork balancing, do it here and not earlier because:
@@ -2434,8 +2517,10 @@ void wake_up_new_task(struct task_struct *p)
*/
set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0));
#endif
+ /* Post initialize new task's util average when its cfs_rq is set */
+ post_init_entity_util_avg(&p->se);
- rq = __task_rq_lock(p);
+ rq = __task_rq_lock(p, &rf);
activate_task(rq, p, 0);
p->on_rq = TASK_ON_RQ_QUEUED;
trace_sched_wakeup_new(p);
@@ -2446,12 +2531,12 @@ void wake_up_new_task(struct task_struct *p)
* Nothing relies on rq->lock after this, so its fine to
* drop it.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, rf.cookie);
p->sched_class->task_woken(rq, p);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, rf.cookie);
}
#endif
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
}
#ifdef CONFIG_PREEMPT_NOTIFIERS
@@ -2713,7 +2798,7 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev)
*/
static __always_inline struct rq *
context_switch(struct rq *rq, struct task_struct *prev,
- struct task_struct *next)
+ struct task_struct *next, struct pin_cookie cookie)
{
struct mm_struct *mm, *oldmm;
@@ -2733,7 +2818,7 @@ context_switch(struct rq *rq, struct task_struct *prev,
atomic_inc(&oldmm->mm_count);
enter_lazy_tlb(oldmm, next);
} else
- switch_mm(oldmm, mm, next);
+ switch_mm_irqs_off(oldmm, mm, next);
if (!prev->mm) {
prev->active_mm = NULL;
@@ -2745,7 +2830,7 @@ context_switch(struct rq *rq, struct task_struct *prev,
* of the scheduler it's an obvious special-case), so we
* do an early lockdep release here:
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
spin_release(&rq->lock.dep_map, 1, _THIS_IP_);
/* Here we just switch the register state and the stack. */
@@ -2867,7 +2952,7 @@ EXPORT_PER_CPU_SYMBOL(kernel_cpustat);
*/
unsigned long long task_sched_runtime(struct task_struct *p)
{
- unsigned long flags;
+ struct rq_flags rf;
struct rq *rq;
u64 ns;
@@ -2887,7 +2972,7 @@ unsigned long long task_sched_runtime(struct task_struct *p)
return p->se.sum_exec_runtime;
#endif
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
/*
* Must be ->curr _and_ ->on_rq. If dequeued, we would
* project cycles that may never be accounted to this
@@ -2898,7 +2983,7 @@ unsigned long long task_sched_runtime(struct task_struct *p)
p->sched_class->update_curr(rq);
}
ns = p->se.sum_exec_runtime;
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return ns;
}
@@ -2918,7 +3003,7 @@ void scheduler_tick(void)
raw_spin_lock(&rq->lock);
update_rq_clock(rq);
curr->sched_class->task_tick(rq, curr, 0);
- update_cpu_load_active(rq);
+ cpu_load_update_active(rq);
calc_global_load_tick(rq);
raw_spin_unlock(&rq->lock);
@@ -2961,6 +3046,20 @@ u64 scheduler_tick_max_deferment(void)
#if defined(CONFIG_PREEMPT) && (defined(CONFIG_DEBUG_PREEMPT) || \
defined(CONFIG_PREEMPT_TRACER))
+/*
+ * If the value passed in is equal to the current preempt count
+ * then we just disabled preemption. Start timing the latency.
+ */
+static inline void preempt_latency_start(int val)
+{
+ if (preempt_count() == val) {
+ unsigned long ip = get_lock_parent_ip();
+#ifdef CONFIG_DEBUG_PREEMPT
+ current->preempt_disable_ip = ip;
+#endif
+ trace_preempt_off(CALLER_ADDR0, ip);
+ }
+}
void preempt_count_add(int val)
{
@@ -2979,17 +3078,21 @@ void preempt_count_add(int val)
DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >=
PREEMPT_MASK - 10);
#endif
- if (preempt_count() == val) {
- unsigned long ip = get_lock_parent_ip();
-#ifdef CONFIG_DEBUG_PREEMPT
- current->preempt_disable_ip = ip;
-#endif
- trace_preempt_off(CALLER_ADDR0, ip);
- }
+ preempt_latency_start(val);
}
EXPORT_SYMBOL(preempt_count_add);
NOKPROBE_SYMBOL(preempt_count_add);
+/*
+ * If the value passed in equals to the current preempt count
+ * then we just enabled preemption. Stop timing the latency.
+ */
+static inline void preempt_latency_stop(int val)
+{
+ if (preempt_count() == val)
+ trace_preempt_on(CALLER_ADDR0, get_lock_parent_ip());
+}
+
void preempt_count_sub(int val)
{
#ifdef CONFIG_DEBUG_PREEMPT
@@ -3006,13 +3109,15 @@ void preempt_count_sub(int val)
return;
#endif
- if (preempt_count() == val)
- trace_preempt_on(CALLER_ADDR0, get_lock_parent_ip());
+ preempt_latency_stop(val);
__preempt_count_sub(val);
}
EXPORT_SYMBOL(preempt_count_sub);
NOKPROBE_SYMBOL(preempt_count_sub);
+#else
+static inline void preempt_latency_start(int val) { }
+static inline void preempt_latency_stop(int val) { }
#endif
/*
@@ -3065,7 +3170,7 @@ static inline void schedule_debug(struct task_struct *prev)
* Pick up the highest-prio task:
*/
static inline struct task_struct *
-pick_next_task(struct rq *rq, struct task_struct *prev)
+pick_next_task(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie)
{
const struct sched_class *class = &fair_sched_class;
struct task_struct *p;
@@ -3076,20 +3181,20 @@ pick_next_task(struct rq *rq, struct task_struct *prev)
*/
if (likely(prev->sched_class == class &&
rq->nr_running == rq->cfs.h_nr_running)) {
- p = fair_sched_class.pick_next_task(rq, prev);
+ p = fair_sched_class.pick_next_task(rq, prev, cookie);
if (unlikely(p == RETRY_TASK))
goto again;
/* assumes fair_sched_class->next == idle_sched_class */
if (unlikely(!p))
- p = idle_sched_class.pick_next_task(rq, prev);
+ p = idle_sched_class.pick_next_task(rq, prev, cookie);
return p;
}
again:
for_each_class(class) {
- p = class->pick_next_task(rq, prev);
+ p = class->pick_next_task(rq, prev, cookie);
if (p) {
if (unlikely(p == RETRY_TASK))
goto again;
@@ -3143,6 +3248,7 @@ static void __sched notrace __schedule(bool preempt)
{
struct task_struct *prev, *next;
unsigned long *switch_count;
+ struct pin_cookie cookie;
struct rq *rq;
int cpu;
@@ -3176,7 +3282,7 @@ static void __sched notrace __schedule(bool preempt)
*/
smp_mb__before_spinlock();
raw_spin_lock(&rq->lock);
- lockdep_pin_lock(&rq->lock);
+ cookie = lockdep_pin_lock(&rq->lock);
rq->clock_skip_update <<= 1; /* promote REQ to ACT */
@@ -3198,7 +3304,7 @@ static void __sched notrace __schedule(bool preempt)
to_wakeup = wq_worker_sleeping(prev);
if (to_wakeup)
- try_to_wake_up_local(to_wakeup);
+ try_to_wake_up_local(to_wakeup, cookie);
}
}
switch_count = &prev->nvcsw;
@@ -3207,7 +3313,7 @@ static void __sched notrace __schedule(bool preempt)
if (task_on_rq_queued(prev))
update_rq_clock(rq);
- next = pick_next_task(rq, prev);
+ next = pick_next_task(rq, prev, cookie);
clear_tsk_need_resched(prev);
clear_preempt_need_resched();
rq->clock_skip_update = 0;
@@ -3218,9 +3324,9 @@ static void __sched notrace __schedule(bool preempt)
++*switch_count;
trace_sched_switch(preempt, prev, next);
- rq = context_switch(rq, prev, next); /* unlocks the rq */
+ rq = context_switch(rq, prev, next, cookie); /* unlocks the rq */
} else {
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
raw_spin_unlock_irq(&rq->lock);
}
@@ -3287,8 +3393,23 @@ void __sched schedule_preempt_disabled(void)
static void __sched notrace preempt_schedule_common(void)
{
do {
+ /*
+ * Because the function tracer can trace preempt_count_sub()
+ * and it also uses preempt_enable/disable_notrace(), if
+ * NEED_RESCHED is set, the preempt_enable_notrace() called
+ * by the function tracer will call this function again and
+ * cause infinite recursion.
+ *
+ * Preemption must be disabled here before the function
+ * tracer can trace. Break up preempt_disable() into two
+ * calls. One to disable preemption without fear of being
+ * traced. The other to still record the preemption latency,
+ * which can also be traced by the function tracer.
+ */
preempt_disable_notrace();
+ preempt_latency_start(1);
__schedule(true);
+ preempt_latency_stop(1);
preempt_enable_no_resched_notrace();
/*
@@ -3340,7 +3461,21 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void)
return;
do {
+ /*
+ * Because the function tracer can trace preempt_count_sub()
+ * and it also uses preempt_enable/disable_notrace(), if
+ * NEED_RESCHED is set, the preempt_enable_notrace() called
+ * by the function tracer will call this function again and
+ * cause infinite recursion.
+ *
+ * Preemption must be disabled here before the function
+ * tracer can trace. Break up preempt_disable() into two
+ * calls. One to disable preemption without fear of being
+ * traced. The other to still record the preemption latency,
+ * which can also be traced by the function tracer.
+ */
preempt_disable_notrace();
+ preempt_latency_start(1);
/*
* Needs preempt disabled in case user_exit() is traced
* and the tracer calls preempt_enable_notrace() causing
@@ -3350,6 +3485,7 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void)
__schedule(true);
exception_exit(prev_ctx);
+ preempt_latency_stop(1);
preempt_enable_no_resched_notrace();
} while (need_resched());
}
@@ -3406,12 +3542,13 @@ EXPORT_SYMBOL(default_wake_function);
void rt_mutex_setprio(struct task_struct *p, int prio)
{
int oldprio, queued, running, queue_flag = DEQUEUE_SAVE | DEQUEUE_MOVE;
- struct rq *rq;
const struct sched_class *prev_class;
+ struct rq_flags rf;
+ struct rq *rq;
BUG_ON(prio > MAX_PRIO);
- rq = __task_rq_lock(p);
+ rq = __task_rq_lock(p, &rf);
/*
* Idle task boosting is a nono in general. There is one
@@ -3487,7 +3624,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
check_class_changed(rq, p, prev_class, oldprio);
out_unlock:
preempt_disable(); /* avoid rq from going away on us */
- __task_rq_unlock(rq);
+ __task_rq_unlock(rq, &rf);
balance_callback(rq);
preempt_enable();
@@ -3497,7 +3634,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
void set_user_nice(struct task_struct *p, long nice)
{
int old_prio, delta, queued;
- unsigned long flags;
+ struct rq_flags rf;
struct rq *rq;
if (task_nice(p) == nice || nice < MIN_NICE || nice > MAX_NICE)
@@ -3506,7 +3643,7 @@ void set_user_nice(struct task_struct *p, long nice)
* We have to be careful, if called from sys_setpriority(),
* the task might be in the middle of scheduling on another CPU.
*/
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
/*
* The RT priorities are set via sched_setscheduler(), but we still
* allow the 'normal' nice value to be set - but as expected
@@ -3537,7 +3674,7 @@ void set_user_nice(struct task_struct *p, long nice)
resched_curr(rq);
}
out_unlock:
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
}
EXPORT_SYMBOL(set_user_nice);
@@ -3834,11 +3971,11 @@ static int __sched_setscheduler(struct task_struct *p,
MAX_RT_PRIO - 1 - attr->sched_priority;
int retval, oldprio, oldpolicy = -1, queued, running;
int new_effective_prio, policy = attr->sched_policy;
- unsigned long flags;
const struct sched_class *prev_class;
- struct rq *rq;
+ struct rq_flags rf;
int reset_on_fork;
int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE;
+ struct rq *rq;
/* may grab non-irq protected spin_locks */
BUG_ON(in_interrupt());
@@ -3933,13 +4070,13 @@ static int __sched_setscheduler(struct task_struct *p,
* To be able to change p->policy safely, the appropriate
* runqueue lock must be held.
*/
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
/*
* Changing the policy of the stop threads its a very bad idea
*/
if (p == rq->stop) {
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return -EINVAL;
}
@@ -3956,7 +4093,7 @@ static int __sched_setscheduler(struct task_struct *p,
goto change;
p->sched_reset_on_fork = reset_on_fork;
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return 0;
}
change:
@@ -3970,7 +4107,7 @@ static int __sched_setscheduler(struct task_struct *p,
if (rt_bandwidth_enabled() && rt_policy(policy) &&
task_group(p)->rt_bandwidth.rt_runtime == 0 &&
!task_group_is_autogroup(task_group(p))) {
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return -EPERM;
}
#endif
@@ -3985,7 +4122,7 @@ static int __sched_setscheduler(struct task_struct *p,
*/
if (!cpumask_subset(span, &p->cpus_allowed) ||
rq->rd->dl_bw.bw == 0) {
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return -EPERM;
}
}
@@ -3995,7 +4132,7 @@ static int __sched_setscheduler(struct task_struct *p,
/* recheck policy now with rq lock held */
if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) {
policy = oldpolicy = -1;
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
goto recheck;
}
@@ -4005,7 +4142,7 @@ static int __sched_setscheduler(struct task_struct *p,
* is available.
*/
if ((dl_policy(policy) || dl_task(p)) && dl_overflow(p, policy, attr)) {
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
return -EBUSY;
}
@@ -4050,7 +4187,7 @@ static int __sched_setscheduler(struct task_struct *p,
check_class_changed(rq, p, prev_class, oldprio);
preempt_disable(); /* avoid rq from going away on us */
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
if (pi)
rt_mutex_adjust_pi(p);
@@ -4903,10 +5040,10 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid,
{
struct task_struct *p;
unsigned int time_slice;
- unsigned long flags;
+ struct rq_flags rf;
+ struct timespec t;
struct rq *rq;
int retval;
- struct timespec t;
if (pid < 0)
return -EINVAL;
@@ -4921,11 +5058,11 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid,
if (retval)
goto out_unlock;
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
time_slice = 0;
if (p->sched_class->get_rr_interval)
time_slice = p->sched_class->get_rr_interval(rq, p);
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
rcu_read_unlock();
jiffies_to_timespec(time_slice, &t);
@@ -5001,7 +5138,8 @@ void show_state_filter(unsigned long state_filter)
touch_all_softlockup_watchdogs();
#ifdef CONFIG_SCHED_DEBUG
- sysrq_sched_debug_show();
+ if (!state_filter)
+ sysrq_sched_debug_show();
#endif
rcu_read_unlock();
/*
@@ -5163,6 +5301,8 @@ int task_can_attach(struct task_struct *p,
#ifdef CONFIG_SMP
+static bool sched_smp_initialized __read_mostly;
+
#ifdef CONFIG_NUMA_BALANCING
/* Migrate current task p to target_cpu */
int migrate_task_to(struct task_struct *p, int target_cpu)
@@ -5188,11 +5328,11 @@ int migrate_task_to(struct task_struct *p, int target_cpu)
*/
void sched_setnuma(struct task_struct *p, int nid)
{
- struct rq *rq;
- unsigned long flags;
bool queued, running;
+ struct rq_flags rf;
+ struct rq *rq;
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
queued = task_on_rq_queued(p);
running = task_current(rq, p);
@@ -5207,7 +5347,7 @@ void sched_setnuma(struct task_struct *p, int nid)
p->sched_class->set_curr_task(rq);
if (queued)
enqueue_task(rq, p, ENQUEUE_RESTORE);
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
}
#endif /* CONFIG_NUMA_BALANCING */
@@ -5223,7 +5363,7 @@ void idle_task_exit(void)
BUG_ON(cpu_online(smp_processor_id()));
if (mm != &init_mm) {
- switch_mm(mm, &init_mm, current);
+ switch_mm_irqs_off(mm, &init_mm, current);
finish_arch_post_lock_switch();
}
mmdrop(mm);
@@ -5271,6 +5411,7 @@ static void migrate_tasks(struct rq *dead_rq)
{
struct rq *rq = dead_rq;
struct task_struct *next, *stop = rq->stop;
+ struct pin_cookie cookie;
int dest_cpu;
/*
@@ -5302,8 +5443,8 @@ static void migrate_tasks(struct rq *dead_rq)
/*
* pick_next_task assumes pinned rq->lock.
*/
- lockdep_pin_lock(&rq->lock);
- next = pick_next_task(rq, &fake_task);
+ cookie = lockdep_pin_lock(&rq->lock);
+ next = pick_next_task(rq, &fake_task, cookie);
BUG_ON(!next);
next->sched_class->put_prev_task(rq, next);
@@ -5316,7 +5457,7 @@ static void migrate_tasks(struct rq *dead_rq)
* because !cpu_active at this point, which means load-balance
* will not interfere. Also, stop-machine.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
raw_spin_unlock(&rq->lock);
raw_spin_lock(&next->pi_lock);
raw_spin_lock(&rq->lock);
@@ -5377,127 +5518,13 @@ static void set_rq_offline(struct rq *rq)
}
}
-/*
- * migration_call - callback that gets triggered when a CPU is added.
- * Here we can start up the necessary migration thread for the new CPU.
- */
-static int
-migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
+static void set_cpu_rq_start_time(unsigned int cpu)
{
- int cpu = (long)hcpu;
- unsigned long flags;
struct rq *rq = cpu_rq(cpu);
- switch (action & ~CPU_TASKS_FROZEN) {
-
- case CPU_UP_PREPARE:
- rq->calc_load_update = calc_load_update;
- account_reset_rq(rq);
- break;
-
- case CPU_ONLINE:
- /* Update our root-domain */
- raw_spin_lock_irqsave(&rq->lock, flags);
- if (rq->rd) {
- BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
-
- set_rq_online(rq);
- }
- raw_spin_unlock_irqrestore(&rq->lock, flags);
- break;
-
-#ifdef CONFIG_HOTPLUG_CPU
- case CPU_DYING:
- sched_ttwu_pending();
- /* Update our root-domain */
- raw_spin_lock_irqsave(&rq->lock, flags);
- if (rq->rd) {
- BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
- set_rq_offline(rq);
- }
- migrate_tasks(rq);
- BUG_ON(rq->nr_running != 1); /* the migration thread */
- raw_spin_unlock_irqrestore(&rq->lock, flags);
- break;
-
- case CPU_DEAD:
- calc_load_migrate(rq);
- break;
-#endif
- }
-
- update_max_interval();
-
- return NOTIFY_OK;
-}
-
-/*
- * Register at high priority so that task migration (migrate_all_tasks)
- * happens before everything else. This has to be lower priority than
- * the notifier in the perf_event subsystem, though.
- */
-static struct notifier_block migration_notifier = {
- .notifier_call = migration_call,
- .priority = CPU_PRI_MIGRATION,
-};
-
-static void set_cpu_rq_start_time(void)
-{
- int cpu = smp_processor_id();
- struct rq *rq = cpu_rq(cpu);
rq->age_stamp = sched_clock_cpu(cpu);
}
-static int sched_cpu_active(struct notifier_block *nfb,
- unsigned long action, void *hcpu)
-{
- int cpu = (long)hcpu;
-
- switch (action & ~CPU_TASKS_FROZEN) {
- case CPU_STARTING:
- set_cpu_rq_start_time();
- return NOTIFY_OK;
-
- case CPU_DOWN_FAILED:
- set_cpu_active(cpu, true);
- return NOTIFY_OK;
-
- default:
- return NOTIFY_DONE;
- }
-}
-
-static int sched_cpu_inactive(struct notifier_block *nfb,
- unsigned long action, void *hcpu)
-{
- switch (action & ~CPU_TASKS_FROZEN) {
- case CPU_DOWN_PREPARE:
- set_cpu_active((long)hcpu, false);
- return NOTIFY_OK;
- default:
- return NOTIFY_DONE;
- }
-}
-
-static int __init migration_init(void)
-{
- void *cpu = (void *)(long)smp_processor_id();
- int err;
-
- /* Initialize migration for the boot CPU */
- err = migration_call(&migration_notifier, CPU_UP_PREPARE, cpu);
- BUG_ON(err == NOTIFY_BAD);
- migration_call(&migration_notifier, CPU_ONLINE, cpu);
- register_cpu_notifier(&migration_notifier);
-
- /* Register cpu active notifiers */
- cpu_notifier(sched_cpu_active, CPU_PRI_SCHED_ACTIVE);
- cpu_notifier(sched_cpu_inactive, CPU_PRI_SCHED_INACTIVE);
-
- return 0;
-}
-early_initcall(migration_init);
-
static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
#ifdef CONFIG_SCHED_DEBUG
@@ -6645,10 +6672,10 @@ static void sched_init_numa(void)
init_numa_topology_type();
}
-static void sched_domains_numa_masks_set(int cpu)
+static void sched_domains_numa_masks_set(unsigned int cpu)
{
- int i, j;
int node = cpu_to_node(cpu);
+ int i, j;
for (i = 0; i < sched_domains_numa_levels; i++) {
for (j = 0; j < nr_node_ids; j++) {
@@ -6658,51 +6685,20 @@ static void sched_domains_numa_masks_set(int cpu)
}
}
-static void sched_domains_numa_masks_clear(int cpu)
+static void sched_domains_numa_masks_clear(unsigned int cpu)
{
int i, j;
+
for (i = 0; i < sched_domains_numa_levels; i++) {
for (j = 0; j < nr_node_ids; j++)
cpumask_clear_cpu(cpu, sched_domains_numa_masks[i][j]);
}
}
-/*
- * Update sched_domains_numa_masks[level][node] array when new cpus
- * are onlined.
- */
-static int sched_domains_numa_masks_update(struct notifier_block *nfb,
- unsigned long action,
- void *hcpu)
-{
- int cpu = (long)hcpu;
-
- switch (action & ~CPU_TASKS_FROZEN) {
- case CPU_ONLINE:
- sched_domains_numa_masks_set(cpu);
- break;
-
- case CPU_DEAD:
- sched_domains_numa_masks_clear(cpu);
- break;
-
- default:
- return NOTIFY_DONE;
- }
-
- return NOTIFY_OK;
-}
#else
-static inline void sched_init_numa(void)
-{
-}
-
-static int sched_domains_numa_masks_update(struct notifier_block *nfb,
- unsigned long action,
- void *hcpu)
-{
- return 0;
-}
+static inline void sched_init_numa(void) { }
+static void sched_domains_numa_masks_set(unsigned int cpu) { }
+static void sched_domains_numa_masks_clear(unsigned int cpu) { }
#endif /* CONFIG_NUMA */
static int __sdt_alloc(const struct cpumask *cpu_map)
@@ -7092,13 +7088,9 @@ static int num_cpus_frozen; /* used to mark begin/end of suspend/resume */
* If we come here as part of a suspend/resume, don't touch cpusets because we
* want to restore it back to its original state upon resume anyway.
*/
-static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action,
- void *hcpu)
+static void cpuset_cpu_active(void)
{
- switch (action) {
- case CPU_ONLINE_FROZEN:
- case CPU_DOWN_FAILED_FROZEN:
-
+ if (cpuhp_tasks_frozen) {
/*
* num_cpus_frozen tracks how many CPUs are involved in suspend
* resume sequence. As long as this is not the last online
@@ -7108,35 +7100,25 @@ static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action,
num_cpus_frozen--;
if (likely(num_cpus_frozen)) {
partition_sched_domains(1, NULL, NULL);
- break;
+ return;
}
-
/*
* This is the last CPU online operation. So fall through and
* restore the original sched domains by considering the
* cpuset configurations.
*/
-
- case CPU_ONLINE:
- cpuset_update_active_cpus(true);
- break;
- default:
- return NOTIFY_DONE;
}
- return NOTIFY_OK;
+ cpuset_update_active_cpus(true);
}
-static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action,
- void *hcpu)
+static int cpuset_cpu_inactive(unsigned int cpu)
{
unsigned long flags;
- long cpu = (long)hcpu;
struct dl_bw *dl_b;
bool overflow;
int cpus;
- switch (action) {
- case CPU_DOWN_PREPARE:
+ if (!cpuhp_tasks_frozen) {
rcu_read_lock_sched();
dl_b = dl_bw_of(cpu);
@@ -7148,19 +7130,120 @@ static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action,
rcu_read_unlock_sched();
if (overflow)
- return notifier_from_errno(-EBUSY);
+ return -EBUSY;
cpuset_update_active_cpus(false);
- break;
- case CPU_DOWN_PREPARE_FROZEN:
+ } else {
num_cpus_frozen++;
partition_sched_domains(1, NULL, NULL);
- break;
- default:
- return NOTIFY_DONE;
}
- return NOTIFY_OK;
+ return 0;
}
+int sched_cpu_activate(unsigned int cpu)
+{
+ struct rq *rq = cpu_rq(cpu);
+ unsigned long flags;
+
+ set_cpu_active(cpu, true);
+
+ if (sched_smp_initialized) {
+ sched_domains_numa_masks_set(cpu);
+ cpuset_cpu_active();
+ }
+
+ /*
+ * Put the rq online, if not already. This happens:
+ *
+ * 1) In the early boot process, because we build the real domains
+ * after all cpus have been brought up.
+ *
+ * 2) At runtime, if cpuset_cpu_active() fails to rebuild the
+ * domains.
+ */
+ raw_spin_lock_irqsave(&rq->lock, flags);
+ if (rq->rd) {
+ BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
+ set_rq_online(rq);
+ }
+ raw_spin_unlock_irqrestore(&rq->lock, flags);
+
+ update_max_interval();
+
+ return 0;
+}
+
+int sched_cpu_deactivate(unsigned int cpu)
+{
+ int ret;
+
+ set_cpu_active(cpu, false);
+ /*
+ * We've cleared cpu_active_mask, wait for all preempt-disabled and RCU
+ * users of this state to go away such that all new such users will
+ * observe it.
+ *
+ * For CONFIG_PREEMPT we have preemptible RCU and its sync_rcu() might
+ * not imply sync_sched(), so wait for both.
+ *
+ * Do sync before park smpboot threads to take care the rcu boost case.
+ */
+ if (IS_ENABLED(CONFIG_PREEMPT))
+ synchronize_rcu_mult(call_rcu, call_rcu_sched);
+ else
+ synchronize_rcu();
+
+ if (!sched_smp_initialized)
+ return 0;
+
+ ret = cpuset_cpu_inactive(cpu);
+ if (ret) {
+ set_cpu_active(cpu, true);
+ return ret;
+ }
+ sched_domains_numa_masks_clear(cpu);
+ return 0;
+}
+
+static void sched_rq_cpu_starting(unsigned int cpu)
+{
+ struct rq *rq = cpu_rq(cpu);
+
+ rq->calc_load_update = calc_load_update;
+ account_reset_rq(rq);
+ update_max_interval();
+}
+
+int sched_cpu_starting(unsigned int cpu)
+{
+ set_cpu_rq_start_time(cpu);
+ sched_rq_cpu_starting(cpu);
+ return 0;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+int sched_cpu_dying(unsigned int cpu)
+{
+ struct rq *rq = cpu_rq(cpu);
+ unsigned long flags;
+
+ /* Handle pending wakeups and then migrate everything off */
+ sched_ttwu_pending();
+ raw_spin_lock_irqsave(&rq->lock, flags);
+ if (rq->rd) {
+ BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
+ set_rq_offline(rq);
+ }
+ migrate_tasks(rq);
+ BUG_ON(rq->nr_running != 1);
+ raw_spin_unlock_irqrestore(&rq->lock, flags);
+ calc_load_migrate(rq);
+ update_max_interval();
+ nohz_balance_exit_idle(cpu);
+ hrtick_clear(rq);
+ return 0;
+}
+#endif
+
void __init sched_init_smp(void)
{
cpumask_var_t non_isolated_cpus;
@@ -7182,12 +7265,6 @@ void __init sched_init_smp(void)
cpumask_set_cpu(smp_processor_id(), non_isolated_cpus);
mutex_unlock(&sched_domains_mutex);
- hotcpu_notifier(sched_domains_numa_masks_update, CPU_PRI_SCHED_ACTIVE);
- hotcpu_notifier(cpuset_cpu_active, CPU_PRI_CPUSET_ACTIVE);
- hotcpu_notifier(cpuset_cpu_inactive, CPU_PRI_CPUSET_INACTIVE);
-
- init_hrtick();
-
/* Move init over to a non-isolated CPU */
if (set_cpus_allowed_ptr(current, non_isolated_cpus) < 0)
BUG();
@@ -7196,7 +7273,16 @@ void __init sched_init_smp(void)
init_sched_rt_class();
init_sched_dl_class();
+ sched_smp_initialized = true;
+}
+
+static int __init migration_init(void)
+{
+ sched_rq_cpu_starting(smp_processor_id());
+ return 0;
}
+early_initcall(migration_init);
+
#else
void __init sched_init_smp(void)
{
@@ -7331,8 +7417,6 @@ void __init sched_init(void)
for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
rq->cpu_load[j] = 0;
- rq->last_load_update_tick = jiffies;
-
#ifdef CONFIG_SMP
rq->sd = NULL;
rq->rd = NULL;
@@ -7351,12 +7435,13 @@ void __init sched_init(void)
rq_attach_root(rq, &def_root_domain);
#ifdef CONFIG_NO_HZ_COMMON
+ rq->last_load_update_tick = jiffies;
rq->nohz_flags = 0;
#endif
#ifdef CONFIG_NO_HZ_FULL
rq->last_sched_tick = 0;
#endif
-#endif
+#endif /* CONFIG_SMP */
init_rq_hrtick(rq);
atomic_set(&rq->nr_iowait, 0);
}
@@ -7394,7 +7479,7 @@ void __init sched_init(void)
if (cpu_isolated_map == NULL)
zalloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT);
idle_thread_set_boot_cpu();
- set_cpu_rq_start_time();
+ set_cpu_rq_start_time(smp_processor_id());
#endif
init_sched_fair_class();
@@ -7639,10 +7724,10 @@ void sched_move_task(struct task_struct *tsk)
{
struct task_group *tg;
int queued, running;
- unsigned long flags;
+ struct rq_flags rf;
struct rq *rq;
- rq = task_rq_lock(tsk, &flags);
+ rq = task_rq_lock(tsk, &rf);
running = task_current(rq, tsk);
queued = task_on_rq_queued(tsk);
@@ -7674,7 +7759,7 @@ void sched_move_task(struct task_struct *tsk)
if (queued)
enqueue_task(rq, tsk, ENQUEUE_RESTORE | ENQUEUE_MOVE);
- task_rq_unlock(rq, tsk, &flags);
+ task_rq_unlock(rq, tsk, &rf);
}
#endif /* CONFIG_CGROUP_SCHED */
@@ -7894,7 +7979,7 @@ static int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk)
static int sched_rt_global_constraints(void)
{
unsigned long flags;
- int i, ret = 0;
+ int i;
raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags);
for_each_possible_cpu(i) {
@@ -7906,7 +7991,7 @@ static int sched_rt_global_constraints(void)
}
raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags);
- return ret;
+ return 0;
}
#endif /* CONFIG_RT_GROUP_SCHED */
diff --git a/kernel/sched/cpuacct.c b/kernel/sched/cpuacct.c
index 4a811203c04a..41f85c4d0938 100644
--- a/kernel/sched/cpuacct.c
+++ b/kernel/sched/cpuacct.c
@@ -25,11 +25,22 @@ enum cpuacct_stat_index {
CPUACCT_STAT_NSTATS,
};
+enum cpuacct_usage_index {
+ CPUACCT_USAGE_USER, /* ... user mode */
+ CPUACCT_USAGE_SYSTEM, /* ... kernel mode */
+
+ CPUACCT_USAGE_NRUSAGE,
+};
+
+struct cpuacct_usage {
+ u64 usages[CPUACCT_USAGE_NRUSAGE];
+};
+
/* track cpu usage of a group of tasks and its child groups */
struct cpuacct {
struct cgroup_subsys_state css;
/* cpuusage holds pointer to a u64-type object on every cpu */
- u64 __percpu *cpuusage;
+ struct cpuacct_usage __percpu *cpuusage;
struct kernel_cpustat __percpu *cpustat;
};
@@ -49,7 +60,7 @@ static inline struct cpuacct *parent_ca(struct cpuacct *ca)
return css_ca(ca->css.parent);
}
-static DEFINE_PER_CPU(u64, root_cpuacct_cpuusage);
+static DEFINE_PER_CPU(struct cpuacct_usage, root_cpuacct_cpuusage);
static struct cpuacct root_cpuacct = {
.cpustat = &kernel_cpustat,
.cpuusage = &root_cpuacct_cpuusage,
@@ -68,7 +79,7 @@ cpuacct_css_alloc(struct cgroup_subsys_state *parent_css)
if (!ca)
goto out;
- ca->cpuusage = alloc_percpu(u64);
+ ca->cpuusage = alloc_percpu(struct cpuacct_usage);
if (!ca->cpuusage)
goto out_free_ca;
@@ -96,20 +107,37 @@ static void cpuacct_css_free(struct cgroup_subsys_state *css)
kfree(ca);
}
-static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu)
+static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu,
+ enum cpuacct_usage_index index)
{
- u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
+ struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
u64 data;
+ /*
+ * We allow index == CPUACCT_USAGE_NRUSAGE here to read
+ * the sum of suages.
+ */
+ BUG_ON(index > CPUACCT_USAGE_NRUSAGE);
+
#ifndef CONFIG_64BIT
/*
* Take rq->lock to make 64-bit read safe on 32-bit platforms.
*/
raw_spin_lock_irq(&cpu_rq(cpu)->lock);
- data = *cpuusage;
+#endif
+
+ if (index == CPUACCT_USAGE_NRUSAGE) {
+ int i = 0;
+
+ data = 0;
+ for (i = 0; i < CPUACCT_USAGE_NRUSAGE; i++)
+ data += cpuusage->usages[i];
+ } else {
+ data = cpuusage->usages[index];
+ }
+
+#ifndef CONFIG_64BIT
raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
-#else
- data = *cpuusage;
#endif
return data;
@@ -117,69 +145,103 @@ static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu)
static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val)
{
- u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
+ struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
+ int i;
#ifndef CONFIG_64BIT
/*
* Take rq->lock to make 64-bit write safe on 32-bit platforms.
*/
raw_spin_lock_irq(&cpu_rq(cpu)->lock);
- *cpuusage = val;
+#endif
+
+ for (i = 0; i < CPUACCT_USAGE_NRUSAGE; i++)
+ cpuusage->usages[i] = val;
+
+#ifndef CONFIG_64BIT
raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
-#else
- *cpuusage = val;
#endif
}
/* return total cpu usage (in nanoseconds) of a group */
-static u64 cpuusage_read(struct cgroup_subsys_state *css, struct cftype *cft)
+static u64 __cpuusage_read(struct cgroup_subsys_state *css,
+ enum cpuacct_usage_index index)
{
struct cpuacct *ca = css_ca(css);
u64 totalcpuusage = 0;
int i;
- for_each_present_cpu(i)
- totalcpuusage += cpuacct_cpuusage_read(ca, i);
+ for_each_possible_cpu(i)
+ totalcpuusage += cpuacct_cpuusage_read(ca, i, index);
return totalcpuusage;
}
+static u64 cpuusage_user_read(struct cgroup_subsys_state *css,
+ struct cftype *cft)
+{
+ return __cpuusage_read(css, CPUACCT_USAGE_USER);
+}
+
+static u64 cpuusage_sys_read(struct cgroup_subsys_state *css,
+ struct cftype *cft)
+{
+ return __cpuusage_read(css, CPUACCT_USAGE_SYSTEM);
+}
+
+static u64 cpuusage_read(struct cgroup_subsys_state *css, struct cftype *cft)
+{
+ return __cpuusage_read(css, CPUACCT_USAGE_NRUSAGE);
+}
+
static int cpuusage_write(struct cgroup_subsys_state *css, struct cftype *cft,
u64 val)
{
struct cpuacct *ca = css_ca(css);
- int err = 0;
- int i;
+ int cpu;
/*
* Only allow '0' here to do a reset.
*/
- if (val) {
- err = -EINVAL;
- goto out;
- }
+ if (val)
+ return -EINVAL;
- for_each_present_cpu(i)
- cpuacct_cpuusage_write(ca, i, 0);
+ for_each_possible_cpu(cpu)
+ cpuacct_cpuusage_write(ca, cpu, 0);
-out:
- return err;
+ return 0;
}
-static int cpuacct_percpu_seq_show(struct seq_file *m, void *V)
+static int __cpuacct_percpu_seq_show(struct seq_file *m,
+ enum cpuacct_usage_index index)
{
struct cpuacct *ca = css_ca(seq_css(m));
u64 percpu;
int i;
- for_each_present_cpu(i) {
- percpu = cpuacct_cpuusage_read(ca, i);
+ for_each_possible_cpu(i) {
+ percpu = cpuacct_cpuusage_read(ca, i, index);
seq_printf(m, "%llu ", (unsigned long long) percpu);
}
seq_printf(m, "\n");
return 0;
}
+static int cpuacct_percpu_user_seq_show(struct seq_file *m, void *V)
+{
+ return __cpuacct_percpu_seq_show(m, CPUACCT_USAGE_USER);
+}
+
+static int cpuacct_percpu_sys_seq_show(struct seq_file *m, void *V)
+{
+ return __cpuacct_percpu_seq_show(m, CPUACCT_USAGE_SYSTEM);
+}
+
+static int cpuacct_percpu_seq_show(struct seq_file *m, void *V)
+{
+ return __cpuacct_percpu_seq_show(m, CPUACCT_USAGE_NRUSAGE);
+}
+
static const char * const cpuacct_stat_desc[] = {
[CPUACCT_STAT_USER] = "user",
[CPUACCT_STAT_SYSTEM] = "system",
@@ -191,7 +253,7 @@ static int cpuacct_stats_show(struct seq_file *sf, void *v)
int cpu;
s64 val = 0;
- for_each_online_cpu(cpu) {
+ for_each_possible_cpu(cpu) {
struct kernel_cpustat *kcpustat = per_cpu_ptr(ca->cpustat, cpu);
val += kcpustat->cpustat[CPUTIME_USER];
val += kcpustat->cpustat[CPUTIME_NICE];
@@ -200,7 +262,7 @@ static int cpuacct_stats_show(struct seq_file *sf, void *v)
seq_printf(sf, "%s %lld\n", cpuacct_stat_desc[CPUACCT_STAT_USER], val);
val = 0;
- for_each_online_cpu(cpu) {
+ for_each_possible_cpu(cpu) {
struct kernel_cpustat *kcpustat = per_cpu_ptr(ca->cpustat, cpu);
val += kcpustat->cpustat[CPUTIME_SYSTEM];
val += kcpustat->cpustat[CPUTIME_IRQ];
@@ -220,10 +282,26 @@ static struct cftype files[] = {
.write_u64 = cpuusage_write,
},
{
+ .name = "usage_user",
+ .read_u64 = cpuusage_user_read,
+ },
+ {
+ .name = "usage_sys",
+ .read_u64 = cpuusage_sys_read,
+ },
+ {
.name = "usage_percpu",
.seq_show = cpuacct_percpu_seq_show,
},
{
+ .name = "usage_percpu_user",
+ .seq_show = cpuacct_percpu_user_seq_show,
+ },
+ {
+ .name = "usage_percpu_sys",
+ .seq_show = cpuacct_percpu_sys_seq_show,
+ },
+ {
.name = "stat",
.seq_show = cpuacct_stats_show,
},
@@ -238,10 +316,17 @@ static struct cftype files[] = {
void cpuacct_charge(struct task_struct *tsk, u64 cputime)
{
struct cpuacct *ca;
+ int index = CPUACCT_USAGE_SYSTEM;
+ struct pt_regs *regs = task_pt_regs(tsk);
+
+ if (regs && user_mode(regs))
+ index = CPUACCT_USAGE_USER;
rcu_read_lock();
+
for (ca = task_ca(tsk); ca; ca = parent_ca(ca))
- *this_cpu_ptr(ca->cpuusage) += cputime;
+ this_cpu_ptr(ca->cpuusage)->usages[index] += cputime;
+
rcu_read_unlock();
}
diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
index 5a75b08cfd85..5be58820465c 100644
--- a/kernel/sched/cpudeadline.c
+++ b/kernel/sched/cpudeadline.c
@@ -103,10 +103,10 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
const struct sched_dl_entity *dl_se = &p->dl;
if (later_mask &&
- cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed)) {
+ cpumask_and(later_mask, cp->free_cpus, tsk_cpus_allowed(p))) {
best_cpu = cpumask_any(later_mask);
goto out;
- } else if (cpumask_test_cpu(cpudl_maximum(cp), &p->cpus_allowed) &&
+ } else if (cpumask_test_cpu(cpudl_maximum(cp), tsk_cpus_allowed(p)) &&
dl_time_before(dl_se->deadline, cp->elements[0].dl)) {
best_cpu = cpudl_maximum(cp);
if (later_mask)
diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c
index 981fcd7dc394..11e9705bf937 100644
--- a/kernel/sched/cpupri.c
+++ b/kernel/sched/cpupri.c
@@ -103,11 +103,11 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p,
if (skip)
continue;
- if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids)
+ if (cpumask_any_and(tsk_cpus_allowed(p), vec->mask) >= nr_cpu_ids)
continue;
if (lowest_mask) {
- cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask);
+ cpumask_and(lowest_mask, tsk_cpus_allowed(p), vec->mask);
/*
* We have to ensure that we have at least one bit
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 686ec8adf952..fcb7f0217ff4 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -134,7 +134,7 @@ static void inc_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
struct task_struct *p = dl_task_of(dl_se);
- if (p->nr_cpus_allowed > 1)
+ if (tsk_nr_cpus_allowed(p) > 1)
dl_rq->dl_nr_migratory++;
update_dl_migration(dl_rq);
@@ -144,7 +144,7 @@ static void dec_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
struct task_struct *p = dl_task_of(dl_se);
- if (p->nr_cpus_allowed > 1)
+ if (tsk_nr_cpus_allowed(p) > 1)
dl_rq->dl_nr_migratory--;
update_dl_migration(dl_rq);
@@ -591,10 +591,10 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
struct sched_dl_entity,
dl_timer);
struct task_struct *p = dl_task_of(dl_se);
- unsigned long flags;
+ struct rq_flags rf;
struct rq *rq;
- rq = task_rq_lock(p, &flags);
+ rq = task_rq_lock(p, &rf);
/*
* The task might have changed its scheduling policy to something
@@ -670,14 +670,14 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
* Nothing relies on rq->lock after this, so its safe to drop
* rq->lock.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, rf.cookie);
push_dl_task(rq);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, rf.cookie);
}
#endif
unlock:
- task_rq_unlock(rq, p, &flags);
+ task_rq_unlock(rq, p, &rf);
/*
* This can free the task_struct, including this hrtimer, do not touch
@@ -717,10 +717,6 @@ static void update_curr_dl(struct rq *rq)
if (!dl_task(curr) || !on_dl_rq(dl_se))
return;
- /* Kick cpufreq (see the comment in linux/cpufreq.h). */
- if (cpu_of(rq) == smp_processor_id())
- cpufreq_trigger_update(rq_clock(rq));
-
/*
* Consumed budget is computed considering the time as
* observed by schedulable tasks (excluding time spent
@@ -736,6 +732,10 @@ static void update_curr_dl(struct rq *rq)
return;
}
+ /* kick cpufreq (see the comment in linux/cpufreq.h). */
+ if (cpu_of(rq) == smp_processor_id())
+ cpufreq_trigger_update(rq_clock(rq));
+
schedstat_set(curr->se.statistics.exec_max,
max(curr->se.statistics.exec_max, delta_exec));
@@ -966,7 +966,7 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
enqueue_dl_entity(&p->dl, pi_se, flags);
- if (!task_current(rq, p) && p->nr_cpus_allowed > 1)
+ if (!task_current(rq, p) && tsk_nr_cpus_allowed(p) > 1)
enqueue_pushable_dl_task(rq, p);
}
@@ -1040,9 +1040,9 @@ select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags)
* try to make it stay here, it might be important.
*/
if (unlikely(dl_task(curr)) &&
- (curr->nr_cpus_allowed < 2 ||
+ (tsk_nr_cpus_allowed(curr) < 2 ||
!dl_entity_preempt(&p->dl, &curr->dl)) &&
- (p->nr_cpus_allowed > 1)) {
+ (tsk_nr_cpus_allowed(p) > 1)) {
int target = find_later_rq(p);
if (target != -1 &&
@@ -1063,7 +1063,7 @@ static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p)
* Current can't be migrated, useless to reschedule,
* let's hope p can move out.
*/
- if (rq->curr->nr_cpus_allowed == 1 ||
+ if (tsk_nr_cpus_allowed(rq->curr) == 1 ||
cpudl_find(&rq->rd->cpudl, rq->curr, NULL) == -1)
return;
@@ -1071,7 +1071,7 @@ static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p)
* p is migratable, so let's not schedule it and
* see if it is pushed or pulled somewhere else.
*/
- if (p->nr_cpus_allowed != 1 &&
+ if (tsk_nr_cpus_allowed(p) != 1 &&
cpudl_find(&rq->rd->cpudl, p, NULL) != -1)
return;
@@ -1125,7 +1125,8 @@ static struct sched_dl_entity *pick_next_dl_entity(struct rq *rq,
return rb_entry(left, struct sched_dl_entity, rb_node);
}
-struct task_struct *pick_next_task_dl(struct rq *rq, struct task_struct *prev)
+struct task_struct *
+pick_next_task_dl(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie)
{
struct sched_dl_entity *dl_se;
struct task_struct *p;
@@ -1140,9 +1141,9 @@ struct task_struct *pick_next_task_dl(struct rq *rq, struct task_struct *prev)
* disabled avoiding further scheduler activity on it and we're
* being very careful to re-start the picking loop.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
pull_dl_task(rq);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, cookie);
/*
* pull_rt_task() can drop (and re-acquire) rq->lock; this
* means a stop task can slip in, in which case we need to
@@ -1185,7 +1186,7 @@ static void put_prev_task_dl(struct rq *rq, struct task_struct *p)
{
update_curr_dl(rq);
- if (on_dl_rq(&p->dl) && p->nr_cpus_allowed > 1)
+ if (on_dl_rq(&p->dl) && tsk_nr_cpus_allowed(p) > 1)
enqueue_pushable_dl_task(rq, p);
}
@@ -1286,7 +1287,7 @@ static int find_later_rq(struct task_struct *task)
if (unlikely(!later_mask))
return -1;
- if (task->nr_cpus_allowed == 1)
+ if (tsk_nr_cpus_allowed(task) == 1)
return -1;
/*
@@ -1392,7 +1393,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq)
if (double_lock_balance(rq, later_rq)) {
if (unlikely(task_rq(task) != rq ||
!cpumask_test_cpu(later_rq->cpu,
- &task->cpus_allowed) ||
+ tsk_cpus_allowed(task)) ||
task_running(rq, task) ||
!dl_task(task) ||
!task_on_rq_queued(task))) {
@@ -1432,7 +1433,7 @@ static struct task_struct *pick_next_pushable_dl_task(struct rq *rq)
BUG_ON(rq->cpu != task_cpu(p));
BUG_ON(task_current(rq, p));
- BUG_ON(p->nr_cpus_allowed <= 1);
+ BUG_ON(tsk_nr_cpus_allowed(p) <= 1);
BUG_ON(!task_on_rq_queued(p));
BUG_ON(!dl_task(p));
@@ -1471,7 +1472,7 @@ static int push_dl_task(struct rq *rq)
*/
if (dl_task(rq->curr) &&
dl_time_before(next_task->dl.deadline, rq->curr->dl.deadline) &&
- rq->curr->nr_cpus_allowed > 1) {
+ tsk_nr_cpus_allowed(rq->curr) > 1) {
resched_curr(rq);
return 0;
}
@@ -1618,9 +1619,9 @@ static void task_woken_dl(struct rq *rq, struct task_struct *p)
{
if (!task_running(rq, p) &&
!test_tsk_need_resched(rq->curr) &&
- p->nr_cpus_allowed > 1 &&
+ tsk_nr_cpus_allowed(p) > 1 &&
dl_task(rq->curr) &&
- (rq->curr->nr_cpus_allowed < 2 ||
+ (tsk_nr_cpus_allowed(rq->curr) < 2 ||
!dl_entity_preempt(&p->dl, &rq->curr->dl))) {
push_dl_tasks(rq);
}
@@ -1724,7 +1725,7 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
if (task_on_rq_queued(p) && rq->curr != p) {
#ifdef CONFIG_SMP
- if (p->nr_cpus_allowed > 1 && rq->dl.overloaded)
+ if (tsk_nr_cpus_allowed(p) > 1 && rq->dl.overloaded)
queue_push_tasks(rq);
#else
if (dl_task(rq->curr))
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 4fbc3bd5ff60..cf905f655ba1 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -626,15 +626,16 @@ do { \
#undef P
#undef PN
-#ifdef CONFIG_SCHEDSTATS
-#define P(n) SEQ_printf(m, " .%-30s: %d\n", #n, rq->n);
-#define P64(n) SEQ_printf(m, " .%-30s: %Ld\n", #n, rq->n);
-
#ifdef CONFIG_SMP
+#define P64(n) SEQ_printf(m, " .%-30s: %Ld\n", #n, rq->n);
P64(avg_idle);
P64(max_idle_balance_cost);
+#undef P64
#endif
+#ifdef CONFIG_SCHEDSTATS
+#define P(n) SEQ_printf(m, " .%-30s: %d\n", #n, rq->n);
+
if (schedstat_enabled()) {
P(yld_count);
P(sched_count);
@@ -644,7 +645,6 @@ do { \
}
#undef P
-#undef P64
#endif
spin_lock_irqsave(&sched_debug_lock, flags);
print_cfs_stats(m, cpu);
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e7dd0ec169be..218f8e83db73 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -204,7 +204,7 @@ static void __update_inv_weight(struct load_weight *lw)
* OR
* (delta_exec * (weight * lw->inv_weight)) >> WMULT_SHIFT
*
- * Either weight := NICE_0_LOAD and lw \e prio_to_wmult[], in which case
+ * Either weight := NICE_0_LOAD and lw \e sched_prio_to_wmult[], in which case
* we're guaranteed shift stays positive because inv_weight is guaranteed to
* fit 32 bits, and NICE_0_LOAD gives another 10 bits; therefore shift >= 22.
*
@@ -682,17 +682,68 @@ void init_entity_runnable_average(struct sched_entity *se)
sa->period_contrib = 1023;
sa->load_avg = scale_load_down(se->load.weight);
sa->load_sum = sa->load_avg * LOAD_AVG_MAX;
- sa->util_avg = scale_load_down(SCHED_LOAD_SCALE);
- sa->util_sum = sa->util_avg * LOAD_AVG_MAX;
+ /*
+ * At this point, util_avg won't be used in select_task_rq_fair anyway
+ */
+ sa->util_avg = 0;
+ sa->util_sum = 0;
/* when this task enqueue'ed, it will contribute to its cfs_rq's load_avg */
}
+/*
+ * With new tasks being created, their initial util_avgs are extrapolated
+ * based on the cfs_rq's current util_avg:
+ *
+ * util_avg = cfs_rq->util_avg / (cfs_rq->load_avg + 1) * se.load.weight
+ *
+ * However, in many cases, the above util_avg does not give a desired
+ * value. Moreover, the sum of the util_avgs may be divergent, such
+ * as when the series is a harmonic series.
+ *
+ * To solve this problem, we also cap the util_avg of successive tasks to
+ * only 1/2 of the left utilization budget:
+ *
+ * util_avg_cap = (1024 - cfs_rq->avg.util_avg) / 2^n
+ *
+ * where n denotes the nth task.
+ *
+ * For example, a simplest series from the beginning would be like:
+ *
+ * task util_avg: 512, 256, 128, 64, 32, 16, 8, ...
+ * cfs_rq util_avg: 512, 768, 896, 960, 992, 1008, 1016, ...
+ *
+ * Finally, that extrapolated util_avg is clamped to the cap (util_avg_cap)
+ * if util_avg > util_avg_cap.
+ */
+void post_init_entity_util_avg(struct sched_entity *se)
+{
+ struct cfs_rq *cfs_rq = cfs_rq_of(se);
+ struct sched_avg *sa = &se->avg;
+ long cap = (long)(SCHED_CAPACITY_SCALE - cfs_rq->avg.util_avg) / 2;
+
+ if (cap > 0) {
+ if (cfs_rq->avg.util_avg != 0) {
+ sa->util_avg = cfs_rq->avg.util_avg * se->load.weight;
+ sa->util_avg /= (cfs_rq->avg.load_avg + 1);
+
+ if (sa->util_avg > cap)
+ sa->util_avg = cap;
+ } else {
+ sa->util_avg = cap;
+ }
+ sa->util_sum = sa->util_avg * LOAD_AVG_MAX;
+ }
+}
+
static inline unsigned long cfs_rq_runnable_load_avg(struct cfs_rq *cfs_rq);
static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq);
#else
void init_entity_runnable_average(struct sched_entity *se)
{
}
+void post_init_entity_util_avg(struct sched_entity *se)
+{
+}
#endif
/*
@@ -2437,10 +2488,12 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se)
update_load_sub(&cfs_rq->load, se->load.weight);
if (!parent_entity(se))
update_load_sub(&rq_of(cfs_rq)->load, se->load.weight);
+#ifdef CONFIG_SMP
if (entity_is_task(se)) {
account_numa_dequeue(rq_of(cfs_rq), task_of(se));
list_del_init(&se->group_node);
}
+#endif
cfs_rq->nr_running--;
}
@@ -2550,6 +2603,16 @@ static const u32 runnable_avg_yN_sum[] = {
};
/*
+ * Precomputed \Sum y^k { 1<=k<=n, where n%32=0). Values are rolled down to
+ * lower integers. See Documentation/scheduler/sched-avg.txt how these
+ * were generated:
+ */
+static const u32 __accumulated_sum_N32[] = {
+ 0, 23371, 35056, 40899, 43820, 45281,
+ 46011, 46376, 46559, 46650, 46696, 46719,
+};
+
+/*
* Approximate:
* val * y^n, where y^32 ~= 0.5 (~1 scheduling period)
*/
@@ -2597,22 +2660,13 @@ static u32 __compute_runnable_contrib(u64 n)
else if (unlikely(n >= LOAD_AVG_MAX_N))
return LOAD_AVG_MAX;
- /* Compute \Sum k^n combining precomputed values for k^i, \Sum k^j */
- do {
- contrib /= 2; /* y^LOAD_AVG_PERIOD = 1/2 */
- contrib += runnable_avg_yN_sum[LOAD_AVG_PERIOD];
-
- n -= LOAD_AVG_PERIOD;
- } while (n > LOAD_AVG_PERIOD);
-
+ /* Since n < LOAD_AVG_MAX_N, n/LOAD_AVG_PERIOD < 11 */
+ contrib = __accumulated_sum_N32[n/LOAD_AVG_PERIOD];
+ n %= LOAD_AVG_PERIOD;
contrib = decay_load(contrib, n);
return contrib + runnable_avg_yN_sum[n];
}
-#if (SCHED_LOAD_SHIFT - SCHED_LOAD_RESOLUTION) != 10 || SCHED_CAPACITY_SHIFT != 10
-#error "load tracking assumes 2^10 as unit"
-#endif
-
#define cap_scale(v, s) ((v)*(s) >> SCHED_CAPACITY_SHIFT)
/*
@@ -2821,23 +2875,54 @@ static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) {}
static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq);
+static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq)
+{
+ struct rq *rq = rq_of(cfs_rq);
+ int cpu = cpu_of(rq);
+
+ if (cpu == smp_processor_id() && &rq->cfs == cfs_rq) {
+ unsigned long max = rq->cpu_capacity_orig;
+
+ /*
+ * There are a few boundary cases this might miss but it should
+ * get called often enough that that should (hopefully) not be
+ * a real problem -- added to that it only calls on the local
+ * CPU, so if we enqueue remotely we'll miss an update, but
+ * the next tick/schedule should update.
+ *
+ * It will not get called when we go idle, because the idle
+ * thread is a different class (!fair), nor will the utilization
+ * number include things like RT tasks.
+ *
+ * As is, the util number is not freq-invariant (we'd have to
+ * implement arch_scale_freq_capacity() for that).
+ *
+ * See cpu_util().
+ */
+ cpufreq_update_util(rq_clock(rq),
+ min(cfs_rq->avg.util_avg, max), max);
+ }
+}
+
/* Group cfs_rq's load_avg is used for task_h_load and update_cfs_share */
-static inline int update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq)
+static inline int
+update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq)
{
struct sched_avg *sa = &cfs_rq->avg;
- int decayed, removed = 0;
+ int decayed, removed_load = 0, removed_util = 0;
if (atomic_long_read(&cfs_rq->removed_load_avg)) {
s64 r = atomic_long_xchg(&cfs_rq->removed_load_avg, 0);
sa->load_avg = max_t(long, sa->load_avg - r, 0);
sa->load_sum = max_t(s64, sa->load_sum - r * LOAD_AVG_MAX, 0);
- removed = 1;
+ removed_load = 1;
}
if (atomic_long_read(&cfs_rq->removed_util_avg)) {
long r = atomic_long_xchg(&cfs_rq->removed_util_avg, 0);
sa->util_avg = max_t(long, sa->util_avg - r, 0);
sa->util_sum = max_t(s32, sa->util_sum - r * LOAD_AVG_MAX, 0);
+ removed_util = 1;
}
decayed = __update_load_avg(now, cpu_of(rq_of(cfs_rq)), sa,
@@ -2848,7 +2933,10 @@ static inline int update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq)
cfs_rq->load_last_update_time_copy = sa->last_update_time;
#endif
- return decayed || removed;
+ if (update_freq && (decayed || removed_util))
+ cfs_rq_util_change(cfs_rq);
+
+ return decayed || removed_load;
}
/* Update task and its cfs_rq load average */
@@ -2867,31 +2955,8 @@ static inline void update_load_avg(struct sched_entity *se, int update_tg)
se->on_rq * scale_load_down(se->load.weight),
cfs_rq->curr == se, NULL);
- if (update_cfs_rq_load_avg(now, cfs_rq) && update_tg)
+ if (update_cfs_rq_load_avg(now, cfs_rq, true) && update_tg)
update_tg_load_avg(cfs_rq, 0);
-
- if (cpu == smp_processor_id() && &rq->cfs == cfs_rq) {
- unsigned long max = rq->cpu_capacity_orig;
-
- /*
- * There are a few boundary cases this might miss but it should
- * get called often enough that that should (hopefully) not be
- * a real problem -- added to that it only calls on the local
- * CPU, so if we enqueue remotely we'll miss an update, but
- * the next tick/schedule should update.
- *
- * It will not get called when we go idle, because the idle
- * thread is a different class (!fair), nor will the utilization
- * number include things like RT tasks.
- *
- * As is, the util number is not freq-invariant (we'd have to
- * implement arch_scale_freq_capacity() for that).
- *
- * See cpu_util().
- */
- cpufreq_update_util(rq_clock(rq),
- min(cfs_rq->avg.util_avg, max), max);
- }
}
static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
@@ -2919,6 +2984,8 @@ static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s
cfs_rq->avg.load_sum += se->avg.load_sum;
cfs_rq->avg.util_avg += se->avg.util_avg;
cfs_rq->avg.util_sum += se->avg.util_sum;
+
+ cfs_rq_util_change(cfs_rq);
}
static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
@@ -2931,6 +2998,8 @@ static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s
cfs_rq->avg.load_sum = max_t(s64, cfs_rq->avg.load_sum - se->avg.load_sum, 0);
cfs_rq->avg.util_avg = max_t(long, cfs_rq->avg.util_avg - se->avg.util_avg, 0);
cfs_rq->avg.util_sum = max_t(s32, cfs_rq->avg.util_sum - se->avg.util_sum, 0);
+
+ cfs_rq_util_change(cfs_rq);
}
/* Add the load generated by se into cfs_rq's load average */
@@ -2948,7 +3017,7 @@ enqueue_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
cfs_rq->curr == se, NULL);
}
- decayed = update_cfs_rq_load_avg(now, cfs_rq);
+ decayed = update_cfs_rq_load_avg(now, cfs_rq, !migrated);
cfs_rq->runnable_load_avg += sa->load_avg;
cfs_rq->runnable_load_sum += sa->load_sum;
@@ -3185,20 +3254,61 @@ static inline void check_schedstat_required(void)
#endif
}
+
+/*
+ * MIGRATION
+ *
+ * dequeue
+ * update_curr()
+ * update_min_vruntime()
+ * vruntime -= min_vruntime
+ *
+ * enqueue
+ * update_curr()
+ * update_min_vruntime()
+ * vruntime += min_vruntime
+ *
+ * this way the vruntime transition between RQs is done when both
+ * min_vruntime are up-to-date.
+ *
+ * WAKEUP (remote)
+ *
+ * ->migrate_task_rq_fair() (p->state == TASK_WAKING)
+ * vruntime -= min_vruntime
+ *
+ * enqueue
+ * update_curr()
+ * update_min_vruntime()
+ * vruntime += min_vruntime
+ *
+ * this way we don't have the most up-to-date min_vruntime on the originating
+ * CPU and an up-to-date min_vruntime on the destination CPU.
+ */
+
static void
enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
{
+ bool renorm = !(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_MIGRATED);
+ bool curr = cfs_rq->curr == se;
+
/*
- * Update the normalized vruntime before updating min_vruntime
- * through calling update_curr().
+ * If we're the current task, we must renormalise before calling
+ * update_curr().
*/
- if (!(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_WAKING))
+ if (renorm && curr)
se->vruntime += cfs_rq->min_vruntime;
+ update_curr(cfs_rq);
+
/*
- * Update run-time statistics of the 'current'.
+ * Otherwise, renormalise after, such that we're placed at the current
+ * moment in time, instead of some random moment in the past. Being
+ * placed in the past could significantly boost this task to the
+ * fairness detriment of existing tasks.
*/
- update_curr(cfs_rq);
+ if (renorm && !curr)
+ se->vruntime += cfs_rq->min_vruntime;
+
enqueue_entity_load_avg(cfs_rq, se);
account_entity_enqueue(cfs_rq, se);
update_cfs_shares(cfs_rq);
@@ -3214,7 +3324,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
update_stats_enqueue(cfs_rq, se);
check_spread(cfs_rq, se);
}
- if (se != cfs_rq->curr)
+ if (!curr)
__enqueue_entity(cfs_rq, se);
se->on_rq = 1;
@@ -4422,7 +4532,7 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
}
#ifdef CONFIG_SMP
-
+#ifdef CONFIG_NO_HZ_COMMON
/*
* per rq 'load' arrray crap; XXX kill this.
*/
@@ -4488,13 +4598,13 @@ decay_load_missed(unsigned long load, unsigned long missed_updates, int idx)
}
return load;
}
+#endif /* CONFIG_NO_HZ_COMMON */
/**
- * __update_cpu_load - update the rq->cpu_load[] statistics
+ * __cpu_load_update - update the rq->cpu_load[] statistics
* @this_rq: The rq to update statistics for
* @this_load: The current load
* @pending_updates: The number of missed updates
- * @active: !0 for NOHZ_FULL
*
* Update rq->cpu_load[] statistics. This function is usually called every
* scheduler tick (TICK_NSEC).
@@ -4523,12 +4633,12 @@ decay_load_missed(unsigned long load, unsigned long missed_updates, int idx)
* load[i]_n = (1 - 1/2^i)^n * load[i]_0
*
* see decay_load_misses(). For NOHZ_FULL we get to subtract and add the extra
- * term. See the @active paramter.
+ * term.
*/
-static void __update_cpu_load(struct rq *this_rq, unsigned long this_load,
- unsigned long pending_updates, int active)
+static void cpu_load_update(struct rq *this_rq, unsigned long this_load,
+ unsigned long pending_updates)
{
- unsigned long tickless_load = active ? this_rq->cpu_load[0] : 0;
+ unsigned long __maybe_unused tickless_load = this_rq->cpu_load[0];
int i, scale;
this_rq->nr_load_updates++;
@@ -4541,6 +4651,7 @@ static void __update_cpu_load(struct rq *this_rq, unsigned long this_load,
/* scale is effectively 1 << i now, and >> i divides by scale */
old_load = this_rq->cpu_load[i];
+#ifdef CONFIG_NO_HZ_COMMON
old_load = decay_load_missed(old_load, pending_updates - 1, i);
if (tickless_load) {
old_load -= decay_load_missed(tickless_load, pending_updates - 1, i);
@@ -4551,6 +4662,7 @@ static void __update_cpu_load(struct rq *this_rq, unsigned long this_load,
*/
old_load += tickless_load;
}
+#endif
new_load = this_load;
/*
* Round up the averaging division if load is increasing. This
@@ -4573,10 +4685,23 @@ static unsigned long weighted_cpuload(const int cpu)
}
#ifdef CONFIG_NO_HZ_COMMON
-static void __update_cpu_load_nohz(struct rq *this_rq,
- unsigned long curr_jiffies,
- unsigned long load,
- int active)
+/*
+ * There is no sane way to deal with nohz on smp when using jiffies because the
+ * cpu doing the jiffies update might drift wrt the cpu doing the jiffy reading
+ * causing off-by-one errors in observed deltas; {0,2} instead of {1,1}.
+ *
+ * Therefore we need to avoid the delta approach from the regular tick when
+ * possible since that would seriously skew the load calculation. This is why we
+ * use cpu_load_update_periodic() for CPUs out of nohz. However we'll rely on
+ * jiffies deltas for updates happening while in nohz mode (idle ticks, idle
+ * loop exit, nohz_idle_balance, nohz full exit...)
+ *
+ * This means we might still be one tick off for nohz periods.
+ */
+
+static void cpu_load_update_nohz(struct rq *this_rq,
+ unsigned long curr_jiffies,
+ unsigned long load)
{
unsigned long pending_updates;
@@ -4588,28 +4713,15 @@ static void __update_cpu_load_nohz(struct rq *this_rq,
* In the NOHZ_FULL case, we were non-idle, we should consider
* its weighted load.
*/
- __update_cpu_load(this_rq, load, pending_updates, active);
+ cpu_load_update(this_rq, load, pending_updates);
}
}
/*
- * There is no sane way to deal with nohz on smp when using jiffies because the
- * cpu doing the jiffies update might drift wrt the cpu doing the jiffy reading
- * causing off-by-one errors in observed deltas; {0,2} instead of {1,1}.
- *
- * Therefore we cannot use the delta approach from the regular tick since that
- * would seriously skew the load calculation. However we'll make do for those
- * updates happening while idle (nohz_idle_balance) or coming out of idle
- * (tick_nohz_idle_exit).
- *
- * This means we might still be one tick off for nohz periods.
- */
-
-/*
* Called from nohz_idle_balance() to update the load ratings before doing the
* idle balance.
*/
-static void update_cpu_load_idle(struct rq *this_rq)
+static void cpu_load_update_idle(struct rq *this_rq)
{
/*
* bail if there's load or we're actually up-to-date.
@@ -4617,38 +4729,71 @@ static void update_cpu_load_idle(struct rq *this_rq)
if (weighted_cpuload(cpu_of(this_rq)))
return;
- __update_cpu_load_nohz(this_rq, READ_ONCE(jiffies), 0, 0);
+ cpu_load_update_nohz(this_rq, READ_ONCE(jiffies), 0);
}
/*
- * Called from tick_nohz_idle_exit() -- try and fix up the ticks we missed.
+ * Record CPU load on nohz entry so we know the tickless load to account
+ * on nohz exit. cpu_load[0] happens then to be updated more frequently
+ * than other cpu_load[idx] but it should be fine as cpu_load readers
+ * shouldn't rely into synchronized cpu_load[*] updates.
*/
-void update_cpu_load_nohz(int active)
+void cpu_load_update_nohz_start(void)
{
struct rq *this_rq = this_rq();
+
+ /*
+ * This is all lockless but should be fine. If weighted_cpuload changes
+ * concurrently we'll exit nohz. And cpu_load write can race with
+ * cpu_load_update_idle() but both updater would be writing the same.
+ */
+ this_rq->cpu_load[0] = weighted_cpuload(cpu_of(this_rq));
+}
+
+/*
+ * Account the tickless load in the end of a nohz frame.
+ */
+void cpu_load_update_nohz_stop(void)
+{
unsigned long curr_jiffies = READ_ONCE(jiffies);
- unsigned long load = active ? weighted_cpuload(cpu_of(this_rq)) : 0;
+ struct rq *this_rq = this_rq();
+ unsigned long load;
if (curr_jiffies == this_rq->last_load_update_tick)
return;
+ load = weighted_cpuload(cpu_of(this_rq));
raw_spin_lock(&this_rq->lock);
- __update_cpu_load_nohz(this_rq, curr_jiffies, load, active);
+ update_rq_clock(this_rq);
+ cpu_load_update_nohz(this_rq, curr_jiffies, load);
raw_spin_unlock(&this_rq->lock);
}
-#endif /* CONFIG_NO_HZ */
+#else /* !CONFIG_NO_HZ_COMMON */
+static inline void cpu_load_update_nohz(struct rq *this_rq,
+ unsigned long curr_jiffies,
+ unsigned long load) { }
+#endif /* CONFIG_NO_HZ_COMMON */
+
+static void cpu_load_update_periodic(struct rq *this_rq, unsigned long load)
+{
+#ifdef CONFIG_NO_HZ_COMMON
+ /* See the mess around cpu_load_update_nohz(). */
+ this_rq->last_load_update_tick = READ_ONCE(jiffies);
+#endif
+ cpu_load_update(this_rq, load, 1);
+}
/*
* Called from scheduler_tick()
*/
-void update_cpu_load_active(struct rq *this_rq)
+void cpu_load_update_active(struct rq *this_rq)
{
unsigned long load = weighted_cpuload(cpu_of(this_rq));
- /*
- * See the mess around update_cpu_load_idle() / update_cpu_load_nohz().
- */
- this_rq->last_load_update_tick = jiffies;
- __update_cpu_load(this_rq, load, 1, 1);
+
+ if (tick_nohz_tick_stopped())
+ cpu_load_update_nohz(this_rq, READ_ONCE(jiffies), load);
+ else
+ cpu_load_update_periodic(this_rq, load);
}
/*
@@ -4706,46 +4851,6 @@ static unsigned long cpu_avg_load_per_task(int cpu)
return 0;
}
-static void record_wakee(struct task_struct *p)
-{
- /*
- * Rough decay (wiping) for cost saving, don't worry
- * about the boundary, really active task won't care
- * about the loss.
- */
- if (time_after(jiffies, current->wakee_flip_decay_ts + HZ)) {
- current->wakee_flips >>= 1;
- current->wakee_flip_decay_ts = jiffies;
- }
-
- if (current->last_wakee != p) {
- current->last_wakee = p;
- current->wakee_flips++;
- }
-}
-
-static void task_waking_fair(struct task_struct *p)
-{
- struct sched_entity *se = &p->se;
- struct cfs_rq *cfs_rq = cfs_rq_of(se);
- u64 min_vruntime;
-
-#ifndef CONFIG_64BIT
- u64 min_vruntime_copy;
-
- do {
- min_vruntime_copy = cfs_rq->min_vruntime_copy;
- smp_rmb();
- min_vruntime = cfs_rq->min_vruntime;
- } while (min_vruntime != min_vruntime_copy);
-#else
- min_vruntime = cfs_rq->min_vruntime;
-#endif
-
- se->vruntime -= min_vruntime;
- record_wakee(p);
-}
-
#ifdef CONFIG_FAIR_GROUP_SCHED
/*
* effective_load() calculates the load change as seen from the root_task_group
@@ -4861,17 +4966,39 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg)
#endif
+static void record_wakee(struct task_struct *p)
+{
+ /*
+ * Only decay a single time; tasks that have less then 1 wakeup per
+ * jiffy will not have built up many flips.
+ */
+ if (time_after(jiffies, current->wakee_flip_decay_ts + HZ)) {
+ current->wakee_flips >>= 1;
+ current->wakee_flip_decay_ts = jiffies;
+ }
+
+ if (current->last_wakee != p) {
+ current->last_wakee = p;
+ current->wakee_flips++;
+ }
+}
+
/*
* Detect M:N waker/wakee relationships via a switching-frequency heuristic.
+ *
* A waker of many should wake a different task than the one last awakened
- * at a frequency roughly N times higher than one of its wakees. In order
- * to determine whether we should let the load spread vs consolodating to
- * shared cache, we look for a minimum 'flip' frequency of llc_size in one
- * partner, and a factor of lls_size higher frequency in the other. With
- * both conditions met, we can be relatively sure that the relationship is
- * non-monogamous, with partner count exceeding socket size. Waker/wakee
- * being client/server, worker/dispatcher, interrupt source or whatever is
- * irrelevant, spread criteria is apparent partner count exceeds socket size.
+ * at a frequency roughly N times higher than one of its wakees.
+ *
+ * In order to determine whether we should let the load spread vs consolidating
+ * to shared cache, we look for a minimum 'flip' frequency of llc_size in one
+ * partner, and a factor of lls_size higher frequency in the other.
+ *
+ * With both conditions met, we can be relatively sure that the relationship is
+ * non-monogamous, with partner count exceeding socket size.
+ *
+ * Waker/wakee being client/server, worker/dispatcher, interrupt source or
+ * whatever is irrelevant, spread criteria is apparent partner count exceeds
+ * socket size.
*/
static int wake_wide(struct task_struct *p)
{
@@ -5176,8 +5303,10 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
int want_affine = 0;
int sync = wake_flags & WF_SYNC;
- if (sd_flag & SD_BALANCE_WAKE)
+ if (sd_flag & SD_BALANCE_WAKE) {
+ record_wakee(p);
want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
+ }
rcu_read_lock();
for_each_domain(cpu, tmp) {
@@ -5257,6 +5386,32 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
static void migrate_task_rq_fair(struct task_struct *p)
{
/*
+ * As blocked tasks retain absolute vruntime the migration needs to
+ * deal with this by subtracting the old and adding the new
+ * min_vruntime -- the latter is done by enqueue_entity() when placing
+ * the task on the new runqueue.
+ */
+ if (p->state == TASK_WAKING) {
+ struct sched_entity *se = &p->se;
+ struct cfs_rq *cfs_rq = cfs_rq_of(se);
+ u64 min_vruntime;
+
+#ifndef CONFIG_64BIT
+ u64 min_vruntime_copy;
+
+ do {
+ min_vruntime_copy = cfs_rq->min_vruntime_copy;
+ smp_rmb();
+ min_vruntime = cfs_rq->min_vruntime;
+ } while (min_vruntime != min_vruntime_copy);
+#else
+ min_vruntime = cfs_rq->min_vruntime;
+#endif
+
+ se->vruntime -= min_vruntime;
+ }
+
+ /*
* We are supposed to update the task to "current" time, then its up to date
* and ready to go to new CPU/cfs_rq. But we have difficulty in getting
* what current time is, so simply throw away the out-of-date time. This
@@ -5439,7 +5594,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_
}
static struct task_struct *
-pick_next_task_fair(struct rq *rq, struct task_struct *prev)
+pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie)
{
struct cfs_rq *cfs_rq = &rq->cfs;
struct sched_entity *se;
@@ -5552,9 +5707,9 @@ pick_next_task_fair(struct rq *rq, struct task_struct *prev)
* further scheduler activity on it and we're being very careful to
* re-start the picking loop.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
new_tasks = idle_balance(rq);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, cookie);
/*
* Because idle_balance() releases (and re-acquires) rq->lock, it is
* possible for any higher priority task to appear. In that case we
@@ -5653,7 +5808,7 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp
* W_i,0 = \Sum_j w_i,j (2)
*
* Where w_i,j is the weight of the j-th runnable task on cpu i. This weight
- * is derived from the nice value as per prio_to_weight[].
+ * is derived from the nice value as per sched_prio_to_weight[].
*
* The weight average is an exponential decay average of the instantaneous
* weight:
@@ -6155,7 +6310,7 @@ static void update_blocked_averages(int cpu)
if (throttled_hierarchy(cfs_rq))
continue;
- if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq))
+ if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq, true))
update_tg_load_avg(cfs_rq, 0);
}
raw_spin_unlock_irqrestore(&rq->lock, flags);
@@ -6216,7 +6371,7 @@ static inline void update_blocked_averages(int cpu)
raw_spin_lock_irqsave(&rq->lock, flags);
update_rq_clock(rq);
- update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq);
+ update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq, true);
raw_spin_unlock_irqrestore(&rq->lock, flags);
}
@@ -6625,6 +6780,9 @@ static bool update_sd_pick_busiest(struct lb_env *env,
if (!(env->sd->flags & SD_ASYM_PACKING))
return true;
+ /* No ASYM_PACKING if target cpu is already busy */
+ if (env->idle == CPU_NOT_IDLE)
+ return true;
/*
* ASYM_PACKING needs to move all the work to the lowest
* numbered CPUs in the group, therefore mark all groups
@@ -6634,7 +6792,8 @@ static bool update_sd_pick_busiest(struct lb_env *env,
if (!sds->busiest)
return true;
- if (group_first_cpu(sds->busiest) > group_first_cpu(sg))
+ /* Prefer to move from highest possible cpu's work */
+ if (group_first_cpu(sds->busiest) < group_first_cpu(sg))
return true;
}
@@ -6780,6 +6939,9 @@ static int check_asym_packing(struct lb_env *env, struct sd_lb_stats *sds)
if (!(env->sd->flags & SD_ASYM_PACKING))
return 0;
+ if (env->idle == CPU_NOT_IDLE)
+ return 0;
+
if (!sds->busiest)
return 0;
@@ -6888,9 +7050,10 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
}
/*
- * In the presence of smp nice balancing, certain scenarios can have
- * max load less than avg load(as we skip the groups at or below
- * its cpu_capacity, while calculating max_load..)
+ * Avg load of busiest sg can be less and avg load of local sg can
+ * be greater than avg load across all sgs of sd because avg load
+ * factors in sg capacity and sgs with smaller group_type are
+ * skipped when updating the busiest sg:
*/
if (busiest->avg_load <= sds->avg_load ||
local->avg_load >= sds->avg_load) {
@@ -6903,11 +7066,12 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
*/
if (busiest->group_type == group_overloaded &&
local->group_type == group_overloaded) {
- load_above_capacity = busiest->sum_nr_running *
- SCHED_LOAD_SCALE;
- if (load_above_capacity > busiest->group_capacity)
+ load_above_capacity = busiest->sum_nr_running * SCHED_CAPACITY_SCALE;
+ if (load_above_capacity > busiest->group_capacity) {
load_above_capacity -= busiest->group_capacity;
- else
+ load_above_capacity *= NICE_0_LOAD;
+ load_above_capacity /= busiest->group_capacity;
+ } else
load_above_capacity = ~0UL;
}
@@ -6915,9 +7079,8 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
* We're trying to get all the cpus to the average_load, so we don't
* want to push ourselves above the average load, nor do we wish to
* reduce the max loaded cpu below the average load. At the same time,
- * we also don't want to reduce the group load below the group capacity
- * (so that we can implement power-savings policies etc). Thus we look
- * for the minimum possible imbalance.
+ * we also don't want to reduce the group load below the group
+ * capacity. Thus we look for the minimum possible imbalance.
*/
max_pull = min(busiest->avg_load - sds->avg_load, load_above_capacity);
@@ -6941,10 +7104,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
/**
* find_busiest_group - Returns the busiest group within the sched_domain
- * if there is an imbalance. If there isn't an imbalance, and
- * the user has opted for power-savings, it returns a group whose
- * CPUs can be put to idle by rebalancing those tasks elsewhere, if
- * such a group exists.
+ * if there is an imbalance.
*
* Also calculates the amount of weighted load which should be moved
* to restore balance.
@@ -6952,9 +7112,6 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
* @env: The load balancing environment.
*
* Return: - The busiest group if imbalance exists.
- * - If no imbalance and user has opted for power-savings balance,
- * return the least loaded group whose CPUs can be
- * put to idle by rebalancing its tasks onto our group.
*/
static struct sched_group *find_busiest_group(struct lb_env *env)
{
@@ -6972,8 +7129,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
busiest = &sds.busiest_stat;
/* ASYM feature bypasses nice load balance check */
- if ((env->idle == CPU_IDLE || env->idle == CPU_NEWLY_IDLE) &&
- check_asym_packing(env, &sds))
+ if (check_asym_packing(env, &sds))
return sds.busiest;
/* There is no busy sibling group to pull tasks from */
@@ -7398,10 +7554,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
&busiest->active_balance_work);
}
- /*
- * We've kicked active balancing, reset the failure
- * counter.
- */
+ /* We've kicked active balancing, force task migration. */
sd->nr_balance_failed = sd->cache_nice_tries+1;
}
} else
@@ -7636,10 +7789,13 @@ static int active_load_balance_cpu_stop(void *data)
schedstat_inc(sd, alb_count);
p = detach_one_task(&env);
- if (p)
+ if (p) {
schedstat_inc(sd, alb_pushed);
- else
+ /* Active balancing done, reset the failure counter. */
+ sd->nr_balance_failed = 0;
+ } else {
schedstat_inc(sd, alb_failed);
+ }
}
rcu_read_unlock();
out_unlock:
@@ -7710,7 +7866,7 @@ static void nohz_balancer_kick(void)
return;
}
-static inline void nohz_balance_exit_idle(int cpu)
+void nohz_balance_exit_idle(unsigned int cpu)
{
if (unlikely(test_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu)))) {
/*
@@ -7783,18 +7939,6 @@ void nohz_balance_enter_idle(int cpu)
atomic_inc(&nohz.nr_cpus);
set_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu));
}
-
-static int sched_ilb_notifier(struct notifier_block *nfb,
- unsigned long action, void *hcpu)
-{
- switch (action & ~CPU_TASKS_FROZEN) {
- case CPU_DYING:
- nohz_balance_exit_idle(smp_processor_id());
- return NOTIFY_OK;
- default:
- return NOTIFY_DONE;
- }
-}
#endif
static DEFINE_SPINLOCK(balancing);
@@ -7956,7 +8100,7 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle)
if (time_after_eq(jiffies, rq->next_balance)) {
raw_spin_lock_irq(&rq->lock);
update_rq_clock(rq);
- update_cpu_load_idle(rq);
+ cpu_load_update_idle(rq);
raw_spin_unlock_irq(&rq->lock);
rebalance_domains(rq, CPU_IDLE);
}
@@ -8381,6 +8525,7 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
init_cfs_rq(cfs_rq);
init_tg_cfs_entry(tg, cfs_rq, se, i, parent->se[i]);
init_entity_runnable_average(se);
+ post_init_entity_util_avg(se);
}
return 1;
@@ -8537,7 +8682,6 @@ const struct sched_class fair_sched_class = {
.rq_online = rq_online_fair,
.rq_offline = rq_offline_fair,
- .task_waking = task_waking_fair,
.task_dead = task_dead_fair,
.set_cpus_allowed = set_cpus_allowed_common,
#endif
@@ -8599,7 +8743,6 @@ __init void init_sched_fair_class(void)
#ifdef CONFIG_NO_HZ_COMMON
nohz.next_balance = jiffies;
zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT);
- cpu_notifier(sched_ilb_notifier, 0);
#endif
#endif /* SMP */
diff --git a/kernel/sched/idle_task.c b/kernel/sched/idle_task.c
index 47ce94931f1b..2ce5458bbe1d 100644
--- a/kernel/sched/idle_task.c
+++ b/kernel/sched/idle_task.c
@@ -24,7 +24,7 @@ static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int fl
}
static struct task_struct *
-pick_next_task_idle(struct rq *rq, struct task_struct *prev)
+pick_next_task_idle(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie)
{
put_prev_task(rq, prev);
diff --git a/kernel/sched/loadavg.c b/kernel/sched/loadavg.c
index ef7159012cf3..b0b93fd33af9 100644
--- a/kernel/sched/loadavg.c
+++ b/kernel/sched/loadavg.c
@@ -99,10 +99,13 @@ long calc_load_fold_active(struct rq *this_rq)
static unsigned long
calc_load(unsigned long load, unsigned long exp, unsigned long active)
{
- load *= exp;
- load += active * (FIXED_1 - exp);
- load += 1UL << (FSHIFT - 1);
- return load >> FSHIFT;
+ unsigned long newload;
+
+ newload = load * exp + active * (FIXED_1 - exp);
+ if (active >= load)
+ newload += FIXED_1-1;
+
+ return newload / FIXED_1;
}
#ifdef CONFIG_NO_HZ_COMMON
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index ec4f538d4396..d5690b722691 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -334,7 +334,7 @@ static void inc_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
rt_rq = &rq_of_rt_rq(rt_rq)->rt;
rt_rq->rt_nr_total++;
- if (p->nr_cpus_allowed > 1)
+ if (tsk_nr_cpus_allowed(p) > 1)
rt_rq->rt_nr_migratory++;
update_rt_migration(rt_rq);
@@ -351,7 +351,7 @@ static void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
rt_rq = &rq_of_rt_rq(rt_rq)->rt;
rt_rq->rt_nr_total--;
- if (p->nr_cpus_allowed > 1)
+ if (tsk_nr_cpus_allowed(p) > 1)
rt_rq->rt_nr_migratory--;
update_rt_migration(rt_rq);
@@ -953,14 +953,14 @@ static void update_curr_rt(struct rq *rq)
if (curr->sched_class != &rt_sched_class)
return;
- /* Kick cpufreq (see the comment in linux/cpufreq.h). */
- if (cpu_of(rq) == smp_processor_id())
- cpufreq_trigger_update(rq_clock(rq));
-
delta_exec = rq_clock_task(rq) - curr->se.exec_start;
if (unlikely((s64)delta_exec <= 0))
return;
+ /* Kick cpufreq (see the comment in linux/cpufreq.h). */
+ if (cpu_of(rq) == smp_processor_id())
+ cpufreq_trigger_update(rq_clock(rq));
+
schedstat_set(curr->se.statistics.exec_max,
max(curr->se.statistics.exec_max, delta_exec));
@@ -1324,7 +1324,7 @@ enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags)
enqueue_rt_entity(rt_se, flags);
- if (!task_current(rq, p) && p->nr_cpus_allowed > 1)
+ if (!task_current(rq, p) && tsk_nr_cpus_allowed(p) > 1)
enqueue_pushable_task(rq, p);
}
@@ -1413,7 +1413,7 @@ select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags)
* will have to sort it out.
*/
if (curr && unlikely(rt_task(curr)) &&
- (curr->nr_cpus_allowed < 2 ||
+ (tsk_nr_cpus_allowed(curr) < 2 ||
curr->prio <= p->prio)) {
int target = find_lowest_rq(p);
@@ -1437,7 +1437,7 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p)
* Current can't be migrated, useless to reschedule,
* let's hope p can move out.
*/
- if (rq->curr->nr_cpus_allowed == 1 ||
+ if (tsk_nr_cpus_allowed(rq->curr) == 1 ||
!cpupri_find(&rq->rd->cpupri, rq->curr, NULL))
return;
@@ -1445,7 +1445,7 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p)
* p is migratable, so let's not schedule it and
* see if it is pushed or pulled somewhere else.
*/
- if (p->nr_cpus_allowed != 1
+ if (tsk_nr_cpus_allowed(p) != 1
&& cpupri_find(&rq->rd->cpupri, p, NULL))
return;
@@ -1524,7 +1524,7 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq)
}
static struct task_struct *
-pick_next_task_rt(struct rq *rq, struct task_struct *prev)
+pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie)
{
struct task_struct *p;
struct rt_rq *rt_rq = &rq->rt;
@@ -1536,9 +1536,9 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev)
* disabled avoiding further scheduler activity on it and we're
* being very careful to re-start the picking loop.
*/
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, cookie);
pull_rt_task(rq);
- lockdep_pin_lock(&rq->lock);
+ lockdep_repin_lock(&rq->lock, cookie);
/*
* pull_rt_task() can drop (and re-acquire) rq->lock; this
* means a dl or stop task can slip in, in which case we need
@@ -1579,7 +1579,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
* The previous task needs to be made eligible for pushing
* if it is still active
*/
- if (on_rt_rq(&p->rt) && p->nr_cpus_allowed > 1)
+ if (on_rt_rq(&p->rt) && tsk_nr_cpus_allowed(p) > 1)
enqueue_pushable_task(rq, p);
}
@@ -1629,7 +1629,7 @@ static int find_lowest_rq(struct task_struct *task)
if (unlikely(!lowest_mask))
return -1;
- if (task->nr_cpus_allowed == 1)
+ if (tsk_nr_cpus_allowed(task) == 1)
return -1; /* No other targets possible */
if (!cpupri_find(&task_rq(task)->rd->cpupri, task, lowest_mask))
@@ -1762,7 +1762,7 @@ static struct task_struct *pick_next_pushable_task(struct rq *rq)
BUG_ON(rq->cpu != task_cpu(p));
BUG_ON(task_current(rq, p));
- BUG_ON(p->nr_cpus_allowed <= 1);
+ BUG_ON(tsk_nr_cpus_allowed(p) <= 1);
BUG_ON(!task_on_rq_queued(p));
BUG_ON(!rt_task(p));
@@ -2122,9 +2122,9 @@ static void task_woken_rt(struct rq *rq, struct task_struct *p)
{
if (!task_running(rq, p) &&
!test_tsk_need_resched(rq->curr) &&
- p->nr_cpus_allowed > 1 &&
+ tsk_nr_cpus_allowed(p) > 1 &&
(dl_task(rq->curr) || rt_task(rq->curr)) &&
- (rq->curr->nr_cpus_allowed < 2 ||
+ (tsk_nr_cpus_allowed(rq->curr) < 2 ||
rq->curr->prio <= p->prio))
push_rt_tasks(rq);
}
@@ -2197,7 +2197,7 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
*/
if (task_on_rq_queued(p) && rq->curr != p) {
#ifdef CONFIG_SMP
- if (p->nr_cpus_allowed > 1 && rq->rt.overloaded)
+ if (tsk_nr_cpus_allowed(p) > 1 && rq->rt.overloaded)
queue_push_tasks(rq);
#else
if (p->prio < rq->curr->prio)
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index ec2e8d23527e..e51145e76807 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -31,9 +31,9 @@ extern void calc_global_load_tick(struct rq *this_rq);
extern long calc_load_fold_active(struct rq *this_rq);
#ifdef CONFIG_SMP
-extern void update_cpu_load_active(struct rq *this_rq);
+extern void cpu_load_update_active(struct rq *this_rq);
#else
-static inline void update_cpu_load_active(struct rq *this_rq) { }
+static inline void cpu_load_update_active(struct rq *this_rq) { }
#endif
/*
@@ -49,25 +49,32 @@ static inline void update_cpu_load_active(struct rq *this_rq) { }
* and does not change the user-interface for setting shares/weights.
*
* We increase resolution only if we have enough bits to allow this increased
- * resolution (i.e. BITS_PER_LONG > 32). The costs for increasing resolution
- * when BITS_PER_LONG <= 32 are pretty high and the returns do not justify the
- * increased costs.
+ * resolution (i.e. 64bit). The costs for increasing resolution when 32bit are
+ * pretty high and the returns do not justify the increased costs.
+ *
+ * Really only required when CONFIG_FAIR_GROUP_SCHED is also set, but to
+ * increase coverage and consistency always enable it on 64bit platforms.
*/
-#if 0 /* BITS_PER_LONG > 32 -- currently broken: it increases power usage under light load */
-# define SCHED_LOAD_RESOLUTION 10
-# define scale_load(w) ((w) << SCHED_LOAD_RESOLUTION)
-# define scale_load_down(w) ((w) >> SCHED_LOAD_RESOLUTION)
+#ifdef CONFIG_64BIT
+# define NICE_0_LOAD_SHIFT (SCHED_FIXEDPOINT_SHIFT + SCHED_FIXEDPOINT_SHIFT)
+# define scale_load(w) ((w) << SCHED_FIXEDPOINT_SHIFT)
+# define scale_load_down(w) ((w) >> SCHED_FIXEDPOINT_SHIFT)
#else
-# define SCHED_LOAD_RESOLUTION 0
+# define NICE_0_LOAD_SHIFT (SCHED_FIXEDPOINT_SHIFT)
# define scale_load(w) (w)
# define scale_load_down(w) (w)
#endif
-#define SCHED_LOAD_SHIFT (10 + SCHED_LOAD_RESOLUTION)
-#define SCHED_LOAD_SCALE (1L << SCHED_LOAD_SHIFT)
-
-#define NICE_0_LOAD SCHED_LOAD_SCALE
-#define NICE_0_SHIFT SCHED_LOAD_SHIFT
+/*
+ * Task weight (visible to users) and its load (invisible to users) have
+ * independent resolution, but they should be well calibrated. We use
+ * scale_load() and scale_load_down(w) to convert between them. The
+ * following must be true:
+ *
+ * scale_load(sched_prio_to_weight[USER_PRIO(NICE_TO_PRIO(0))]) == NICE_0_LOAD
+ *
+ */
+#define NICE_0_LOAD (1L << NICE_0_LOAD_SHIFT)
/*
* Single value that decides SCHED_DEADLINE internal math precision.
@@ -585,11 +592,13 @@ struct rq {
#endif
#define CPU_LOAD_IDX_MAX 5
unsigned long cpu_load[CPU_LOAD_IDX_MAX];
- unsigned long last_load_update_tick;
#ifdef CONFIG_NO_HZ_COMMON
+#ifdef CONFIG_SMP
+ unsigned long last_load_update_tick;
+#endif /* CONFIG_SMP */
u64 nohz_stamp;
unsigned long nohz_flags;
-#endif
+#endif /* CONFIG_NO_HZ_COMMON */
#ifdef CONFIG_NO_HZ_FULL
unsigned long last_sched_tick;
#endif
@@ -854,7 +863,7 @@ DECLARE_PER_CPU(struct sched_domain *, sd_asym);
struct sched_group_capacity {
atomic_t ref;
/*
- * CPU capacity of this group, SCHED_LOAD_SCALE being max capacity
+ * CPU capacity of this group, SCHED_CAPACITY_SCALE being max capacity
* for a single CPU.
*/
unsigned int capacity;
@@ -1159,7 +1168,7 @@ extern const u32 sched_prio_to_wmult[40];
*
* ENQUEUE_HEAD - place at front of runqueue (tail if not specified)
* ENQUEUE_REPLENISH - CBS (replenish runtime and postpone deadline)
- * ENQUEUE_WAKING - sched_class::task_waking was called
+ * ENQUEUE_MIGRATED - the task was migrated during wakeup
*
*/
@@ -1174,9 +1183,9 @@ extern const u32 sched_prio_to_wmult[40];
#define ENQUEUE_HEAD 0x08
#define ENQUEUE_REPLENISH 0x10
#ifdef CONFIG_SMP
-#define ENQUEUE_WAKING 0x20
+#define ENQUEUE_MIGRATED 0x20
#else
-#define ENQUEUE_WAKING 0x00
+#define ENQUEUE_MIGRATED 0x00
#endif
#define RETRY_TASK ((void *)-1UL)
@@ -1200,14 +1209,14 @@ struct sched_class {
* tasks.
*/
struct task_struct * (*pick_next_task) (struct rq *rq,
- struct task_struct *prev);
+ struct task_struct *prev,
+ struct pin_cookie cookie);
void (*put_prev_task) (struct rq *rq, struct task_struct *p);
#ifdef CONFIG_SMP
int (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags);
void (*migrate_task_rq)(struct task_struct *p);
- void (*task_waking) (struct task_struct *task);
void (*task_woken) (struct rq *this_rq, struct task_struct *task);
void (*set_cpus_allowed)(struct task_struct *p,
@@ -1313,6 +1322,7 @@ extern void init_dl_task_timer(struct sched_dl_entity *dl_se);
unsigned long to_ratio(u64 period, u64 runtime);
extern void init_entity_runnable_average(struct sched_entity *se);
+extern void post_init_entity_util_avg(struct sched_entity *se);
#ifdef CONFIG_NO_HZ_FULL
extern bool sched_can_stop_tick(struct rq *rq);
@@ -1448,86 +1458,32 @@ static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { }
static inline void sched_avg_update(struct rq *rq) { }
#endif
-/*
- * __task_rq_lock - lock the rq @p resides on.
- */
-static inline struct rq *__task_rq_lock(struct task_struct *p)
- __acquires(rq->lock)
-{
- struct rq *rq;
-
- lockdep_assert_held(&p->pi_lock);
-
- for (;;) {
- rq = task_rq(p);
- raw_spin_lock(&rq->lock);
- if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) {
- lockdep_pin_lock(&rq->lock);
- return rq;
- }
- raw_spin_unlock(&rq->lock);
-
- while (unlikely(task_on_rq_migrating(p)))
- cpu_relax();
- }
-}
+struct rq_flags {
+ unsigned long flags;
+ struct pin_cookie cookie;
+};
-/*
- * task_rq_lock - lock p->pi_lock and lock the rq @p resides on.
- */
-static inline struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags)
+struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf)
+ __acquires(rq->lock);
+struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf)
__acquires(p->pi_lock)
- __acquires(rq->lock)
-{
- struct rq *rq;
-
- for (;;) {
- raw_spin_lock_irqsave(&p->pi_lock, *flags);
- rq = task_rq(p);
- raw_spin_lock(&rq->lock);
- /*
- * move_queued_task() task_rq_lock()
- *
- * ACQUIRE (rq->lock)
- * [S] ->on_rq = MIGRATING [L] rq = task_rq()
- * WMB (__set_task_cpu()) ACQUIRE (rq->lock);
- * [S] ->cpu = new_cpu [L] task_rq()
- * [L] ->on_rq
- * RELEASE (rq->lock)
- *
- * If we observe the old cpu in task_rq_lock, the acquire of
- * the old rq->lock will fully serialize against the stores.
- *
- * If we observe the new cpu in task_rq_lock, the acquire will
- * pair with the WMB to ensure we must then also see migrating.
- */
- if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) {
- lockdep_pin_lock(&rq->lock);
- return rq;
- }
- raw_spin_unlock(&rq->lock);
- raw_spin_unlock_irqrestore(&p->pi_lock, *flags);
-
- while (unlikely(task_on_rq_migrating(p)))
- cpu_relax();
- }
-}
+ __acquires(rq->lock);
-static inline void __task_rq_unlock(struct rq *rq)
+static inline void __task_rq_unlock(struct rq *rq, struct rq_flags *rf)
__releases(rq->lock)
{
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, rf->cookie);
raw_spin_unlock(&rq->lock);
}
static inline void
-task_rq_unlock(struct rq *rq, struct task_struct *p, unsigned long *flags)
+task_rq_unlock(struct rq *rq, struct task_struct *p, struct rq_flags *rf)
__releases(rq->lock)
__releases(p->pi_lock)
{
- lockdep_unpin_lock(&rq->lock);
+ lockdep_unpin_lock(&rq->lock, rf->cookie);
raw_spin_unlock(&rq->lock);
- raw_spin_unlock_irqrestore(&p->pi_lock, *flags);
+ raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags);
}
#ifdef CONFIG_SMP
@@ -1743,6 +1699,10 @@ enum rq_nohz_flag_bits {
};
#define nohz_flags(cpu) (&cpu_rq(cpu)->nohz_flags)
+
+extern void nohz_balance_exit_idle(unsigned int cpu);
+#else
+static inline void nohz_balance_exit_idle(unsigned int cpu) { }
#endif
#ifdef CONFIG_IRQ_TIME_ACCOUNTING
diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
index cbc67da10954..604297a08b3a 100644
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -24,7 +24,7 @@ check_preempt_curr_stop(struct rq *rq, struct task_struct *p, int flags)
}
static struct task_struct *
-pick_next_task_stop(struct rq *rq, struct task_struct *prev)
+pick_next_task_stop(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie)
{
struct task_struct *stop = rq->stop;
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 58e3310c9b21..31872bc53bc4 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -776,6 +776,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
if (!ts->tick_stopped) {
nohz_balance_enter_idle(cpu);
calc_load_enter_idle();
+ cpu_load_update_nohz_start();
ts->last_tick = hrtimer_get_expires(&ts->sched_timer);
ts->tick_stopped = 1;
@@ -802,11 +803,11 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
return tick;
}
-static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now, int active)
+static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now)
{
/* Update jiffies first */
tick_do_update_jiffies64(now);
- update_cpu_load_nohz(active);
+ cpu_load_update_nohz_stop();
calc_load_exit_idle();
touch_softlockup_watchdog_sched();
@@ -833,7 +834,7 @@ static void tick_nohz_full_update_tick(struct tick_sched *ts)
if (can_stop_full_tick(ts))
tick_nohz_stop_sched_tick(ts, ktime_get(), cpu);
else if (ts->tick_stopped)
- tick_nohz_restart_sched_tick(ts, ktime_get(), 1);
+ tick_nohz_restart_sched_tick(ts, ktime_get());
#endif
}
@@ -1024,7 +1025,7 @@ void tick_nohz_idle_exit(void)
tick_nohz_stop_idle(ts, now);
if (ts->tick_stopped) {
- tick_nohz_restart_sched_tick(ts, now, 0);
+ tick_nohz_restart_sched_tick(ts, now);
tick_nohz_account_idle_ticks(ts);
}
diff --git a/mm/mmu_context.c b/mm/mmu_context.c
index f802c2d216a7..6f4d27c5bb32 100644
--- a/mm/mmu_context.c
+++ b/mm/mmu_context.c
@@ -4,9 +4,9 @@
*/
#include <linux/mm.h>
+#include <linux/sched.h>
#include <linux/mmu_context.h>
#include <linux/export.h>
-#include <linux/sched.h>
#include <asm/mmu_context.h>