membarrier: Document scheduler barrier requirements
PATCH v2 1/1 membarrier: Document scheduler barrier requirements
Document the membarrier requirement on having a full memory barrier in __schedule() after coming from user-space, before storing to rq->curr. It is provided by smp_mb__before_spinlock() in __schedule().
Document that membarrier requires a full barrier on transition from kernel thread to userspace thread, which skips the call to switch_mm(). We currently have an implicit barrier from atomic_dec_and_test() in mmdrop() that ensures this.
The x86 switch_mm_irqs_off() full barrier is currently provided by many cpumask update operations as well as load_cr3(). Document that load_cr3() is providing this barrier.
[ Rebased on top of linux-rcu for-mingo branch.
Applies on top of "membarrier: Provide expedited private command". ]
--- arch/x86/mm/tlb.c | 3 +++ include/linux/sched/mm.h | 4 ++++ kernel/sched/core.c | 9 +++++++++ 3 files changed, 16 insertions(+)
arch/x86/mm/tlb.c +3 -0
@@ -133,6 +133,9 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,* and neither LOCK nor MFENCE orders them. * Fortunately, load_cr3() is serializing and gives the * ordering guarantee we need. +* +* This full barrier is also required by the membarrier +* system call. */ load_cr3 (next->pgd);
include/linux/sched/mm.h +4 -0
@@ -38,6 +38,10 @@ static inline void mmgrab(struct mm_struct *mm)extern void __mmdrop (struct mm_struct *);static inline void mmdrop (struct mm_struct *mm) { +/* +* The implicit full barrier implied by atomic_dec_and_test is +* required by the membarrier system call. +*/ if (unlikely (atomic_dec_and_test (&mm->mm_count)))__mmdrop (mm); }
kernel/sched/core.c +9 -0
@@ -2654,6 +2654,12 @@ static struct rq *finish_task_switch(struct task_struct *prev)finish_arch_post_lock_switch ();fire_sched_in_preempt_notifiers (current); +/* +* When transitioning from a kernel thread to a userspace +* thread, mmdrop()'s implicit full barrier is required by the +* membarrier system call, because the current active_mm can +* become the current mm without going through switch_mm(). +*/ if (mm)mmdrop (mm);if (unlikely (prev_state ==TASK_DEAD )) {
@@ -3295,6 +3301,9 @@ static void __sched notrace __schedule(bool preempt)* Make sure that signal_pending_state()->signal_pending() below * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) * done by the caller to avoid the race with signal_wake_up(). +* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. */ smp_mb__before_spinlock ();rq_lock (rq, &rf);
Signature · 1 line
1.9.1
Quoted 9 lines · Mathieu Desnoyers
Document the membarrier requirement on having a full memory barrier in
__schedule()after coming from user-space, before storing torq->curr. It is provided bysmp_mb__before_spinlock()in__schedule().Document that membarrier requires a full barrier on transition from kernel thread to userspace thread, which skips the call to
switch_mm(). We currently have an implicit barrier fromatomic_dec_and_test()inmmdrop()that ensures this.
Quoted 6 lines
The x86
switch_mm_irqs_off()full barrier is currently provided by many cpumask update operations as well asload_cr3(). Document thatload_cr3()is providing this barrier.[ Rebased on top of linux-rcu for-mingo branch.
Applies on top of "membarrier: Provide expedited private command". ]
I have queued this for review and testing, thank you!
Thanx, Paul
Quoted 77 lines
Signed-off-by Mathieu Desnoyers <[email protected]>CC Peter Zijlstra <[email protected]>CC Paul E. McKenney <[email protected]>CC Boqun Feng <[email protected]>CC Andrew Hunter <[email protected]>CC Maged Michael <[email protected]>CC Avi Kivity <[email protected]>CC Benjamin Herrenschmidt <[email protected]>CC Paul Mackerras <[email protected]>CC Michael Ellerman <[email protected]>CC Dave Watson <[email protected]>--- arch/x86/mm/tlb.c | 3 +++ include/linux/sched/mm.h | 4 ++++ kernel/sched/core.c | 9 +++++++++ 3 files changed, 16 insertions(+)arch/x86/mm/tlb.c +3
-0@@ -133,6 +133,9 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,* and neither LOCK nor MFENCE orders them. * Fortunately, load_cr3() is serializing and gives the * ordering guarantee we need. +* +* This full barrier is also required by the membarrier +* system call. */ load_cr3 (next->pgd);include/linux/sched/mm.h +4
-0@@ -38,6 +38,10 @@ static inline void mmgrab(struct mm_struct *mm)extern void __mmdrop (struct mm_struct *);static inline void mmdrop (struct mm_struct *mm) { +/* +* The implicit full barrier implied by atomic_dec_and_test is +* required by the membarrier system call. +*/ if (unlikely (atomic_dec_and_test (&mm->mm_count)))__mmdrop (mm); }kernel/sched/core.c +9
-0@@ -2654,6 +2654,12 @@ static struct rq *finish_task_switch(struct task_struct *prev)finish_arch_post_lock_switch ();fire_sched_in_preempt_notifiers (current); +/* +* When transitioning from a kernel thread to a userspace +* thread, mmdrop()'s implicit full barrier is required by the +* membarrier system call, because the current active_mm can +* become the current mm without going through switch_mm(). +*/ if (mm)mmdrop (mm);if (unlikely (prev_state ==TASK_DEAD )) {@@ -3295,6 +3301,9 @@ static void __sched notrace __schedule(bool preempt)* Make sure that signal_pending_state()->signal_pending() below * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) * done by the caller to avoid the race with signal_wake_up(). +* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. */ smp_mb__before_spinlock ();rq_lock (rq, &rf);Signature · 1 line
1.9.1
Peter Zijlstra
maintainerkernel/sched/core.c static void __sched notrace __schedule(bool preempt) · line 3309
Quoted 10 lines · Mathieu Desnoyers
kernel/sched/core.c +3
-0@@ -3295,6 +3301,9 @@ static void __sched notrace __schedule(bool preempt)* Make sure that signal_pending_state()->signal_pending() below * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) * done by the caller to avoid the race with signal_wake_up(). +* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. */ smp_mb__before_spinlock ();rq_lock (rq, &rf);
Merge conflict here, current tip is over to smp_mb__after_spinlock().
Quoted 13 lines · Peter Zijlstra
kernel/sched/core.c static void __sched notrace __schedule(bool preempt) · line 3309
Quoted 10 lines · Mathieu Desnoyers
kernel/sched/core.c +3
-0@@ -3295,6 +3301,9 @@ static void __sched notrace __schedule(bool preempt)* Make sure that signal_pending_state()->signal_pending() below * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) * done by the caller to avoid the race with signal_wake_up(). +* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. */ smp_mb__before_spinlock ();rq_lock (rq, &rf);Merge conflict here, current tip is over to
smp_mb__after_spinlock().
Would some tree other than -rcu be better for that commit?
Thanx, Paul