Skip to contentSkip to conversation

timers/migration: Fix ignored event due to missing CPU update

PATCH Superseded Frederic Weisbecker 3 participants 6 messages Apr 1, 2024 – Apr 5, 2024 High-resolution timers, timer wheel, clockevents

lore
Share & tools
lore

Frederic Weisbecker

author maintainer
#0 lore raw

PATCH 1/1 timers/migration: Fix ignored event due to missing CPU update

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T0
      /         \
    0 (T0)       1 (T1)
    idle         idle

0) The hierarchy has 3 levels. The left part (GRP1:0) is all idle, including CPU 0 and CPU 1 which have a timer each: T0 and T1. They have the same expiry value.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T0
/ \
0 (T0) 1 (T1)
idle idle

1) The migrator in GRP1:1 handles remotely T0. The event is dequeued from the top and T0 executed.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

2) The migrator in GRP1:1 fetches the next timer for CPU 0 and finds none. But it updates the events from its groups, starting with GRP0:0 which now has T1 as its next event. So far so good.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

3) The migrator in GRP1:1 proceeds upward and updates the events in GRP1:0. The child event TGRP0:0 is found queued with the same expiry as before. And therefore it is left unchanged. However the target CPU is not the same but that fact is ignored so TGRP0:0 still points to CPU 0 when it should point to CPU 1.

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T1
      /         \
    0            1 (T1)
    idle         idle

4) The propagation has reached the top level and TGRP1:0, having TGRP0:0 as its first event, also wrongly points to CPU 0. TGRP1:0 is added to the top level group.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

5) The migrator in GRP1:1 dequeues the next event in top level pointing to CPU 0. But since it actually doesn't see any real event in CPU 0, it early returns.

6) T1 is left unhandled until either CPU 0 or CPU 1 wake up.

Some other bad scenario may involve trees with just two levels.

Fix this with checking the CPU, along with the expiry value before considering to early return while updating a queued event.

---
 kernel/time/timer_migration.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
kernel/time/timer_migration.c +4 -1
index c63a0afdcebe..90786bb9a607 100644
@@ -762,8 +762,11 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child,
 	 * queue when the expiry time changed only or when it could be ignored.
 	 */
 	if (timerqueue_node_queued(&evt->nextevt)) {
-		if ((evt->nextevt.expires == nextexp) && !evt->ignore)
+		if ((evt->nextevt.expires == nextexp) && !evt->ignore) {
+			if (evt->cpu != first_childevt->cpu)
+				evt->cpu = first_childevt->cpu;
 			goto check_toplvl;
+		}
 
 		if (!timerqueue_del(&group->events, &evt->nextevt))
 			WRITE_ONCE(group->next_expiry, KTIME_MAX);
Signature · 1 line
2.44.0
author maintainer
#1 lore raw
Quoted 138 lines · Frederic Weisbecker

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T0
      /         \
    0 (T0)       1 (T1)
    idle         idle

0) The hierarchy has 3 levels. The left part (GRP1:0) is all idle, including CPU 0 and CPU 1 which have a timer each: T0 and T1. They have the same expiry value.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T0
/ \
0 (T0) 1 (T1)
idle idle

1) The migrator in GRP1:1 handles remotely T0. The event is dequeued from the top and T0 executed.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

2) The migrator in GRP1:1 fetches the next timer for CPU 0 and finds none. But it updates the events from its groups, starting with GRP0:0 which now has T1 as its next event. So far so good.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

3) The migrator in GRP1:1 proceeds upward and updates the events in GRP1:0. The child event TGRP0:0 is found queued with the same expiry as before. And therefore it is left unchanged. However the target CPU is not the same but that fact is ignored so TGRP0:0 still points to CPU 0 when it should point to CPU 1.

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T1
      /         \
    0            1 (T1)
    idle         idle

4) The propagation has reached the top level and TGRP1:0, having TGRP0:0 as its first event, also wrongly points to CPU 0. TGRP1:0 is added to the top level group.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

5) The migrator in GRP1:1 dequeues the next event in top level pointing to CPU 0. But since it actually doesn't see any real event in CPU 0, it early returns.

Quoted 8 lines

6) T1 is left unhandled until either CPU 0 or CPU 1 wake up.

Some other bad scenario may involve trees with just two levels.

Fix this with checking the CPU, along with the expiry value before considering to early return while updating a queued event.

Fixes 7ee988770326 ("timers: Implement the hierarchical pull model")
#2 lore raw
In reply to Frederic Weisbecker
Quoted 4 lines · Frederic Weisbecker

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

urgh...

Quoted 11 lines
---
 kernel/time/timer_migration.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
kernel/time/timer_migration.c +0 -0
index c63a0afdcebe..90786bb9a607 100644

kernel/time/timer_migration.c bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, · line 767

Quoted 7 lines
kernel/time/timer_migration.c +3 -1
@@ -762,8 +762,11 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child,
 	 * queue when the expiry time changed only or when it could be ignored.
 	 */
 	if (timerqueue_node_queued(&evt->nextevt)) {
-		if ((evt->nextevt.expires == nextexp) && !evt->ignore)
+		if ((evt->nextevt.expires == nextexp) && !evt->ignore) {
+			if (evt->cpu != first_childevt->cpu)
+				evt->cpu = first_childevt->cpu;

Why not just unconditionally overwriting the evt->cpu value here?

kernel/time/timer_migration.c bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, · line 772

Quoted 5 lines
kernel/time/timer_migration.c +1 -0
 			goto check_toplvl;
+		}
 
 		if (!timerqueue_del(&group->events, &evt->nextevt))
 			WRITE_ONCE(group->next_expiry, KTIME_MAX);

Thanks,

Anna-Maria

author maintainer
#3 lore raw

PATCH v2 1/1 timers/migration: Fix ignored event due to missing CPU update

Quoted 15 lines · Anna-Maria Behnsen
Quoted 4 lines · Frederic Weisbecker

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

urgh...

Quoted 6 lines
---
 kernel/time/timer_migration.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
Quoted 14 lines

kernel/time/timer_migration.c bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, · line 767

Quoted 12 lines
kernel/time/timer_migration.c +3 -1
index c63a0afdcebe..90786bb9a607 100644
@@ -762,8 +762,11 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child,
 	 * queue when the expiry time changed only or when it could be ignored.
 	 */
 	if (timerqueue_node_queued(&evt->nextevt)) {
-		if ((evt->nextevt.expires == nextexp) && !evt->ignore)
+		if ((evt->nextevt.expires == nextexp) && !evt->ignore) {
+			if (evt->cpu != first_childevt->cpu)
+				evt->cpu = first_childevt->cpu;

Why not just unconditionally overwriting the evt->cpu value here?

Right! See below:

---
From d038dad7345398a2f6671a3cda98a48805f9eba3 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <[email protected]>
Date: Mon, 1 Apr 2024 23:48:59 +0200
Subject: [PATCH v2] timers/migration: Fix ignored event due to missing CPU update

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T0
      /         \
    0 (T0)       1 (T1)
    idle         idle

0) The hierarchy has 3 levels. The left part (GRP1:0) is all idle, including CPU 0 and CPU 1 which have a timer each: T0 and T1. They have the same expiry value.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T0
/ \
0 (T0) 1 (T1)
idle idle

1) The migrator in GRP1:1 handles remotely T0. The event is dequeued from the top and T0 executed.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

2) The migrator in GRP1:1 fetches the next timer for CPU 0 and finds none. But it updates the events from its groups, starting with GRP0:0 which now has T1 as its next event. So far so good.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

3) The migrator in GRP1:1 proceeds upward and updates the events in GRP1:0. The child event TGRP0:0 is found queued with the same expiry as before. And therefore it is left unchanged. However the target CPU is not the same but that fact is ignored so TGRP0:0 still points to CPU 0 when it should point to CPU 1.

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T1
      /         \
    0            1 (T1)
    idle         idle

4) The propagation has reached the top level and TGRP1:0, having TGRP0:0 as its first event, also wrongly points to CPU 0. TGRP1:0 is added to the top level group.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

5) The migrator in GRP1:1 dequeues the next event in top level pointing to CPU 0. But since it actually doesn't see any real event in CPU 0, it early returns.

6) T1 is left unhandled until either CPU 0 or CPU 1 wake up.

Some other bad scenario may involve trees with just two levels.

Fix this with unconditionally updating the CPU of the child event before considering to early return while updating a queued event with an unchanged expiry value.

Fixes 7ee988770326 ("timers: Implement the hierarchical pull model")
---
 kernel/time/timer_migration.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
kernel/time/timer_migration.c +4 -1
index c63a0afdcebe..e3075e40cb43 100644
@@ -762,8 +762,11 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child,
 	 * queue when the expiry time changed only or when it could be ignored.
 	 */
 	if (timerqueue_node_queued(&evt->nextevt)) {
-		if ((evt->nextevt.expires == nextexp) && !evt->ignore)
+		if ((evt->nextevt.expires == nextexp) && !evt->ignore) {
+			/* Make sure not to miss a new CPU event with the same expiry */
+			evt->cpu = first_childevt->cpu;
 			goto check_toplvl;
+		}
 
 		if (!timerqueue_del(&group->events, &evt->nextevt))
 			WRITE_ONCE(group->next_expiry, KTIME_MAX);
Signature · 1 line
2.44.0
maintainer Reviewed-by on v2 1/1
#4 lore raw
Quoted 179 lines · Frederic Weisbecker
Quoted 15 lines · Anna-Maria Behnsen
Quoted 4 lines · Frederic Weisbecker

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

urgh...

Quoted 6 lines
---
 kernel/time/timer_migration.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
Quoted 14 lines

kernel/time/timer_migration.c bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, · line 767

Quoted 12 lines
kernel/time/timer_migration.c +3 -1
index c63a0afdcebe..90786bb9a607 100644
@@ -762,8 +762,11 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child,
 	 * queue when the expiry time changed only or when it could be ignored.
 	 */
 	if (timerqueue_node_queued(&evt->nextevt)) {
-		if ((evt->nextevt.expires == nextexp) && !evt->ignore)
+		if ((evt->nextevt.expires == nextexp) && !evt->ignore) {
+			if (evt->cpu != first_childevt->cpu)
+				evt->cpu = first_childevt->cpu;

Why not just unconditionally overwriting the evt->cpu value here?

Right! See below:

---
From d038dad7345398a2f6671a3cda98a48805f9eba3 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <[email protected]>
Date: Mon, 1 Apr 2024 23:48:59 +0200
Subject: [PATCH v2] timers/migration: Fix ignored event due to missing CPU update

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T0
      /         \
    0 (T0)       1 (T1)
    idle         idle

0) The hierarchy has 3 levels. The left part (GRP1:0) is all idle, including CPU 0 and CPU 1 which have a timer each: T0 and T1. They have the same expiry value.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T0
/ \
0 (T0) 1 (T1)
idle idle

1) The migrator in GRP1:1 handles remotely T0. The event is dequeued from the top and T0 executed.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

2) The migrator in GRP1:1 fetches the next timer for CPU 0 and finds none. But it updates the events from its groups, starting with GRP0:0 which now has T1 as its next event. So far so good.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

3) The migrator in GRP1:1 proceeds upward and updates the events in GRP1:0. The child event TGRP0:0 is found queued with the same expiry as before. And therefore it is left unchanged. However the target CPU is not the same but that fact is ignored so TGRP0:0 still points to CPU 0 when it should point to CPU 1.

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T1
      /         \
    0            1 (T1)
    idle         idle

4) The propagation has reached the top level and TGRP1:0, having TGRP0:0 as its first event, also wrongly points to CPU 0. TGRP1:0 is added to the top level group.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

5) The migrator in GRP1:1 dequeues the next event in top level pointing to CPU 0. But since it actually doesn't see any real event in CPU 0, it early returns.

6) T1 is left unhandled until either CPU 0 or CPU 1 wake up.

Quoted 8 lines

Some other bad scenario may involve trees with just two levels.

Fix this with unconditionally updating the CPU of the child event before considering to early return while updating a queued event with an unchanged expiry value.

Fixes 7ee988770326 ("timers: Implement the hierarchical pull model")

Thanks,

Anna-Maria

#5 lore raw
In reply to Frederic Weisbecker
The following commit has been merged into the timers/urgent branch of tip:

The following commit has been merged into the timers/urgent branch of tip:

Commit-ID:     61f7fdf8fd00ce33d30ca3fae8d643c0850ce945
Gitweb:        https://git.kernel.org/tip/61f7fdf8fd00ce33d30ca3fae8d643c0850ce945
Author:        Frederic Weisbecker <[email protected]>
AuthorDate:    Mon, 01 Apr 2024 23:48:59 +02:00
Committer:     Thomas Gleixner <[email protected]>
CommitterDate: Fri, 05 Apr 2024 11:05:16 +02:00

timers/migration: Fix ignored event due to missing CPU update

When a group event is updated with its expiry unchanged but a different CPU, that target change may go unnoticed and the event may be propagated up with a stale CPU value. The following depicts a scenario that has been actually observed:

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T0
      /         \
    0 (T0)       1 (T1)
    idle         idle

0) The hierarchy has 3 levels. The left part (GRP1:0) is all idle, including CPU 0 and CPU 1 which have a timer each: T0 and T1. They have the same expiry value.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T0
/ \
0 (T0) 1 (T1)
idle idle

1) The migrator in GRP1:1 handles remotely T0. The event is dequeued from the top and T0 executed.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

2) The migrator in GRP1:1 fetches the next timer for CPU 0 and finds none. But it updates the events from its groups, starting with GRP0:0 which now has T1 as its next event. So far so good.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

3) The migrator in GRP1:1 proceeds upward and updates the events in GRP1:0. The child event TGRP0:0 is found queued with the same expiry as before. And therefore it is left unchanged. However the target CPU is not the same but that fact is ignored so TGRP0:0 still points to CPU 0 when it should point to CPU 1.

                       [GRP2:0]
                   migrator = GRP1:1
                   active   = GRP1:1
                   nextevt  = TGRP1:0 (T0)
                    /              \
               [GRP1:0]           [GRP1:1]
            migrator = NONE       [...]
            active   = NONE
            nextevt  = TGRP0:0 (T0)
            /           \
        [GRP0:0]       [...]
      migrator = NONE
      active   = NONE
      nextevt  = T1
      /         \
    0            1 (T1)
    idle         idle

4) The propagation has reached the top level and TGRP1:0, having TGRP0:0 as its first event, also wrongly points to CPU 0. TGRP1:0 is added to the top level group.

[GRP2:0]
migrator = GRP1:1
active = GRP1:1
nextevt = KTIME_MAX
/ \
[GRP1:0] [GRP1:1]
migrator = NONE [...]
active = NONE
nextevt = TGRP0:0 (T0)
/ \
[GRP0:0] [...]
migrator = NONE
active = NONE
nextevt = T1
/ \
0 1 (T1)
idle idle

5) The migrator in GRP1:1 dequeues the next event in top level pointing to CPU 0. But since it actually doesn't see any real event in CPU 0, it early returns.

6) T1 is left unhandled until either CPU 0 or CPU 1 wake up.

Some other bad scenario may involve trees with just two levels.

Fix this with unconditionally updating the CPU of the child event before considering to early return while updating a queued event with an unchanged expiry value.

Fixes 7ee988770326 ("timers: Implement the hierarchical pull model")
---
 kernel/time/timer_migration.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
kernel/time/timer_migration.c +4 -1
index c63a0af..e3075e4 100644
@@ -762,8 +762,11 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child,
 	 * queue when the expiry time changed only or when it could be ignored.
 	 */
 	if (timerqueue_node_queued(&evt->nextevt)) {
-		if ((evt->nextevt.expires == nextexp) && !evt->ignore)
+		if ((evt->nextevt.expires == nextexp) && !evt->ignore) {
+			/* Make sure not to miss a new CPU event with the same expiry */
+			evt->cpu = first_childevt->cpu;
 			goto check_toplvl;
+		}
 
 		if (!timerqueue_del(&group->events, &evt->nextevt))
 			WRITE_ONCE(group->next_expiry, KTIME_MAX);