membarrier: Document scheduler barrier requirements
PATCH 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. 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 write_cr3(). Document that write_cr3() provides this barrier.
--- arch/x86/mm/tlb.c | 5 +++++ include/linux/sched/mm.h | 4 ++++ kernel/sched/core.c | 9 +++++++++ 3 files changed, 18 insertions(+)
arch/x86/mm/tlb.c +5 -0
@@ -144,6 +144,11 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, }#endif +/* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. +* Writing to CR3 provides that full memory barrier. +*/ if (real_prev == next) {VM_BUG_ON (this_cpu_read (cpu_tlbstate.ctxs[prev_asid].ctx_id) != next->context.ctx_id);
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
@@ -2648,6 +2648,12 @@ 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 )) {
@@ -3289,6 +3295,9 @@ 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. */ rq_lock (rq, &rf);smp_mb__after_spinlock ();
Signature · 1 line
2.11.0
Paul E. McKenney
AppliedQuoted 8 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. We currently have an implicit barrier from
atomic_dec_and_test()inmmdrop()that ensures this.
Quoted 5 lines
The x86
switch_mm_irqs_off()full barrier is currently provided by many cpumask update operations as well aswrite_cr3(). Document thatwrite_cr3()provides this barrier.Signed-off-by Mathieu Desnoyers <[email protected]>
Queued for review, thank you Mathieu!
Thanx, Paul
Quoted 78 lines
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 | 5 +++++ include/linux/sched/mm.h | 4 ++++ kernel/sched/core.c | 9 +++++++++ 3 files changed, 18 insertions(+)arch/x86/mm/tlb.c +5
-0@@ -144,6 +144,11 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, }#endif +/* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. +* Writing to CR3 provides that full memory barrier. +*/ if (real_prev == next) {VM_BUG_ON (this_cpu_read (cpu_tlbstate.ctxs[prev_asid].ctx_id) != next->context.ctx_id);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@@ -2648,6 +2648,12 @@ 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 )) {@@ -3289,6 +3295,9 @@ 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. */ rq_lock (rq, &rf);smp_mb__after_spinlock ();Signature · 1 line
2.11.0
Quoted 35 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. We currently have an implicit barrier from
atomic_dec_and_test()inmmdrop()that ensures this.The x86
switch_mm_irqs_off()full barrier is currently provided by many cpumask update operations as well aswrite_cr3(). Document thatwrite_cr3()provides this barrier.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 | 5 +++++ include/linux/sched/mm.h | 4 ++++ kernel/sched/core.c | 9 +++++++++ 3 files changed, 18 insertions(+)arch/x86/mm/tlb.c +0
-0
arch/x86/mm/tlb.c void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, · line 149
Quoted 6 lines
arch/x86/mm/tlb.c +3
-0@@ -144,6 +144,11 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, }#endif +/* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr.
I'm confused: isn't this case covered by the
rq_lock(rq, &rf); smp_mb__after_spinlock();
below (all archs)? You meant "before returning to user-space, after storing to rq->curr"?
Quoted 10 lines
arch/x86/mm/tlb.c +2
-0+* Writing to CR3 provides that full memory barrier. +*/ if (real_prev == next) {VM_BUG_ON (this_cpu_read (cpu_tlbstate.ctxs[prev_asid].ctx_id) != next->context.ctx_id);include/linux/sched/mm.h +0
-0
include/linux/sched/mm.h static inline void mmgrab(struct mm_struct *mm) · line 43
Quoted 6 lines
include/linux/sched/mm.h +3
-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.
"before returning to user-space, after storing to rq->curr", right?
(your commit says "on transition from kernel thread _to_ userspace"
regarding mmdrop()).
Andrea
Quoted 34 lines
include/linux/sched/mm.h +1
-0+*/ if (unlikely (atomic_dec_and_test (&mm->mm_count)))__mmdrop (mm); }kernel/sched/core.c +9
-0@@ -2648,6 +2648,12 @@ 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 )) {@@ -3289,6 +3295,9 @@ 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. */ rq_lock (rq, &rf);smp_mb__after_spinlock ();Signature · 1 line
2.11.0
Quoted 46 lines · ----- On Sep 18, 2017, at 8:26 PM, Andrea Parri [email protected] wrote:
Quoted 35 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. We currently have an implicit barrier from
atomic_dec_and_test()inmmdrop()that ensures this.The x86
switch_mm_irqs_off()full barrier is currently provided by many cpumask update operations as well aswrite_cr3(). Document thatwrite_cr3()provides this barrier.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 | 5 +++++ include/linux/sched/mm.h | 4 ++++ kernel/sched/core.c | 9 +++++++++ 3 files changed, 18 insertions(+)arch/x86/mm/tlb.c +0
-0arch/x86/mm/tlb.c void switch_mm_irqs_off(struct mm_struct *prev, struct · line 150
Quoted 7 lines
arch/x86/mm/tlb.c +3
-0@@ -144,6 +144,11 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, }#endif +/* +* The membarrier system call requires a full memory barrier +* after coming from user-space, before storing to rq->curr. I'm confused: isn't this case covered by the
Quoted 5 lines
rq_lock(rq, &rf); smp_mb__after_spinlock();below (all archs)? You meant "before returning to user-space, after storing to
rq->curr"?
Right, my bad.
Quoted 12 lines
Quoted 12 lines
arch/x86/mm/tlb.c +2
-0+* Writing to CR3 provides that full memory barrier. +*/ if (real_prev == next) { VM_BUG_ON(this_cpu_read(cpu_tlbstate.ctxs[prev_asid].ctx_id) !=arch/x86/mm/tlb.c +0
-0next->context.ctx_id);include/linux/sched/mm.h +0
-0@@ -38,6 +38,10 @@ static inline void mmgrab(struct mm_struct *mm)
Quoted 8 lines
include/linux/sched/mm.h static inline void mmgrab(struct mm_struct *mm) · line 43
Quoted 6 lines
include/linux/sched/mm.h +3
-0extern 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. "before returning to user-space, after storing to
rq->curr", right?
Yes.
I'll send an updated patch to Paul. Thanks for the review!
Mathieu
Quoted 41 lines
(your commit says "on transition from kernel thread _to_ userspace" regarding
mmdrop()).Andrea
Quoted 34 lines
include/linux/sched/mm.h +1
-0+*/ if (unlikely (atomic_dec_and_test (&mm->mm_count)))__mmdrop (mm); }kernel/sched/core.c +6
-0@@ -2648,6 +2648,12 @@ 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);kernel/sched/core.c +3
-1if (unlikely (prev_state ==TASK_DEAD )) {@@ -3289,6 +3295,9 @@ 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. */ rq_lock (rq, &rf);smp_mb__after_spinlock (); --2.11.0
Signature · 3 lines
Mathieu Desnoyers EfficiOS Inc. http://www.efficios.com