Skip to contentSkip to conversation

membarrier: Document scheduler barrier requirements

PATCH Superseded Mathieu Desnoyers 3 participants 4 messages Sep 19, 2017 – Sep 22, 2017

Subsystems (3)

v3 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

PATCH v3 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__after_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.

Changes since v1:
- Update comments to match reality for code paths which are after storing to rq->curr, before returning to user-space.
Changes since v2:
- Update changelog (smp_mb__before_spinlock -> smp_mb__after_spinlock).

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 | 5 +++++
 kernel/sched/core.c      | 9 +++++++++
 3 files changed, 19 insertions(+)
arch/x86/mm/tlb.c +5 -0
index 1ab3821f9e26..74f94fe4aded 100644
@@ -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
+	 * before returning to user-space, after 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 +5 -0
index 3a19c253bdb1..766cc47c4d7c 100644
@@ -38,6 +38,11 @@ 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.
+	 */
 	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
 		__mmdrop(mm);
 }
kernel/sched/core.c +9 -0
index 18a6966567da..7977b25acf54 100644
@@ -2658,6 +2658,12 @@ 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)) {
@@ -3299,6 +3305,9 @@ 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
#1 lore raw
Quoted 12 lines · Mathieu Desnoyers

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__after_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.

Quoted 7 lines

Changes since v1:
- Update comments to match reality for code paths which are after storing to rq->curr, before returning to user-space.
Changes since v2:
- Update changelog (smp_mb__before_spinlock -> smp_mb__after_spinlock).

I replaced the version that I had queued with this one, thank you!

Thanx, Paul

Quoted 79 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 | 5 +++++
 kernel/sched/core.c      | 9 +++++++++
 3 files changed, 19 insertions(+)
arch/x86/mm/tlb.c +5 -0
index 1ab3821f9e26..74f94fe4aded 100644
@@ -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
+	 * before returning to user-space, after 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 +5 -0
index 3a19c253bdb1..766cc47c4d7c 100644
@@ -38,6 +38,11 @@ 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.
+	 */
 	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
 		__mmdrop(mm);
 }
kernel/sched/core.c +9 -0
index 18a6966567da..7977b25acf54 100644
@@ -2658,6 +2658,12 @@ 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)) {
@@ -3299,6 +3305,9 @@ 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
maintainer
#2 lore raw
In reply to Mathieu Desnoyers
Quoted 39 lines · Mathieu Desnoyers
arch/x86/mm/tlb.c +5 -0
index 1ab3821f9e26..74f94fe4aded 100644
@@ -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
+	 * before returning to user-space, after 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 +5 -0
index 3a19c253bdb1..766cc47c4d7c 100644
@@ -38,6 +38,11 @@ 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.
+	 */
 	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
 		__mmdrop(mm);
 }
kernel/sched/core.c +0 -0
index 18a6966567da..7977b25acf54 100644
@@ -2658,6 +2658,12 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 	finish_arch_post_lock_switch();
 

kernel/sched/core.c static struct rq *finish_task_switch(struct task_struct *prev) · line 2669

Quoted 10 lines
kernel/sched/core.c +6 -0
 	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)) {

I would also put a comment in context_switch() that explains we either pass through switch_mm() or do mmdrop().

And I think that for the weak archs that don't have native RELEASE we actually rely on rq_unlock() for the smp_mb().

So there's 4 schemes:

- switch_mm()/mmdrop() (x86,s390, sparc?)
- finish_lock_switch() (weak, !release)
- switch_to (arm64)
- member arch hook (ppc)

And I don't think that's spelled out clearly enough.

kernel/sched/core.c static void __sched notrace __schedule(bool preempt) · line 3313

Quoted 10 lines
kernel/sched/core.c +3 -0
@@ -3299,6 +3305,9 @@ 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();

Right, this is the only part that's actually trivial :-)

#3 lore raw
Quoted 69 lines · ----- On Sep 21, 2017, at 8:25 AM, Peter Zijlstra [email protected] wrote:
Quoted 50 lines · Mathieu Desnoyers
arch/x86/mm/tlb.c +5 -0
index 1ab3821f9e26..74f94fe4aded 100644
@@ -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
+	 * before returning to user-space, after 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) !=
arch/x86/mm/tlb.c +0 -0
 			  next->context.ctx_id);
include/linux/sched/mm.h +5 -0
index 3a19c253bdb1..766cc47c4d7c 100644
@@ -38,6 +38,11 @@ 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.
+	 */
 	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
 		__mmdrop(mm);
 }
kernel/sched/core.c +6 -0
index 18a6966567da..7977b25acf54 100644
@@ -2658,6 +2658,12 @@ 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);
Quoted 1 line

if (unlikely(prev_state == TASK_DEAD)) {

I would also put a comment in context_switch() that explains we either pass through switch_mm() or do mmdrop().

And I think that for the weak archs that don't have native RELEASE we actually rely on rq_unlock() for the smp_mb().

So there's 4 schemes:

- switch_mm()/mmdrop() (x86,s390, sparc?)
- finish_lock_switch() (weak, !release)
- switch_to (arm64)
- member arch hook (ppc)

And I don't think that's spelled out clearly enough.

Quoted 12 lines

kernel/sched/core.c static void __sched notrace __schedule(bool preempt) · line 3313

Quoted 10 lines
kernel/sched/core.c +3 -0
@@ -3299,6 +3305,9 @@ 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();

Right, this is the only part that's actually trivial :-)

Does something like this work ? (except for tabs vs spaces)

kernel/sched/core.c +18 -10
index 08095bb1cfe6..6254f87645de 100644
@@ -2760,6 +2760,13 @@ context_switch(struct rq *rq, struct task_struct *prev,
         */
        arch_start_context_switch(prev);
 
+       /*
+        * If mm is non-NULL, we pass through switch_mm(). If mm is
+        * NULL, we will pass through mmdrop() in finish_task_switch().
+        * Both of these contain the full memory barrier required by
+        * membarrier after storing to rq->curr, before returning to
+        * user-space.
+        */
        if (!mm) {
                next->active_mm = oldmm;
                mmgrab(oldmm);
@@ -3346,16 +3353,17 @@ static void __sched notrace __schedule(bool preempt)
                /*
                 * The membarrier system call requires each architecture
                 * to have a full memory barrier after updating
-                * rq->curr, before returning to user-space. For TSO
-                * (e.g. x86), the architecture must provide its own
-                * barrier in switch_mm(). For weakly ordered machines
-                * for which spin_unlock() acts as a full memory
-                * barrier, finish_lock_switch() in common code takes
-                * care of this barrier. For weakly ordered machines for
-                * which spin_unlock() acts as a RELEASE barrier (only
-                * arm64 and PowerPC), arm64 has a full barrier in
-                * switch_to(), and PowerPC has a full barrier in
-                * membarrier_arch_sched_in().
+                * rq->curr, before returning to user-space.
+                *
+                * Here are the schemes providing that barrier on the
+                * various architectures:
+                * - mm ? switch_mm() : mmdrop() for x86, s390, sparc,
+                * - finish_lock_switch() for weakly-ordered
+                *   architectures where spin_unlock is a full barrier,
+                * - switch_to() for arm64 (weakly-ordered, spin_unlock
+                *   is a RELEASE barrier),
+                * - membarrier_arch_sched_in() for PowerPC,
+                *   (weakly-ordered, spin_unlock is a RELEASE barrier).
                 */
                ++*switch_count;
 

Thanks,

Mathieu

Signature · 3 lines
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com