Frederic Weisbecker d4b3a4c2aa timers/migration: Fix imbalanced NUMA trees ... [ Upstream commit 5eb579dfd46b4949117ecb0f1ba2f12d3dc9a6f2 ] When a CPU from a new node boots, the old root may happen to be connected to the new root even if their node mismatch, as depicted in the following scenario: 1) CPU 0 boots and creates the first group for node 0. [GRP0:0] node 0 | CPU 0 2) CPU 1 from node 1 boots and creates a new top that corresponds to node 1, but it also connects the old root from node 0 to the new root from node 1 by mistake. [GRP1:0] node 1 / \ / \ [GRP0:0] [GRP0:1] node 0 node 1 | | CPU 0 CPU 1 3) This eventually leads to an imbalanced tree where some node 0 CPUs migrate node 1 timers (and vice versa) way before reaching the crossnode groups, resulting in more frequent remote memory accesses than expected. [GRP2:0] NUMA_NO_NODE / \ [GRP1:0] [GRP1:1] node 1 node 0 / \ | / \ [...] [GRP0:0] [GRP0:1] node 0 node 1 | | CPU 0... CPU 1... A balanced tree should only contain groups having children that belong to the same node: [GRP2:0] NUMA_NO_NODE / \ [GRP1:0] [GRP1:0] node 0 node 1 / \ / \ / \ / \ [GRP0:0] [...] [...] [GRP0:1] node 0 node 1 | | CPU 0... CPU 1... In order to fix this, the hierarchy must be unfolded up to the crossnode level as soon as a node mismatch is detected. For example the stage 2 above should lead to this layout: [GRP2:0] NUMA_NO_NODE / \ [GRP1:0] [GRP1:1] node 0 node 1 / \ / \ [GRP0:0] [GRP0:1] node 0 node 1 | | CPU 0 CPU 1 This means that not only GRP1:0 must be created but also GRP1:1 and GRP2:0 in order to prepare a balanced tree for next CPUs to boot. Fixes: 7ee9887703 ("timers: Implement the hierarchical pull model") Signed-off-by: Frederic Weisbecker <frederic@kernel.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://patch.msgid.link/20251024132536.39841-4-frederic@kernel.org Signed-off-by: Sasha Levin <sashal@kernel.org>		2025-12-18 14:02:47 +01:00
..
alarmtimer.c	time: Fix spelling mistakes in comments	2025-09-21 10:02:02 +02:00
clockevents.c	tick: Do not set device to detached state in tick_shutdown()	2025-09-09 13:39:00 +02:00
clocksource-wdtest.c
clocksource.c	time: Fix spelling mistakes in comments	2025-09-21 10:02:02 +02:00
hrtimer.c	Updates for the time(rs) core subsystem:	2025-09-30 16:09:27 -07:00
itimer.c	timers/itimer: Avoid direct access to hrtimer clockbase	2025-09-09 12:27:17 +02:00
jiffies.c
Kconfig
Makefile	time: Build generic update_vsyscall() only with generic time vDSO	2025-09-04 11:23:50 +02:00
namespace.c	namespace-6.18-rc1	2025-09-29 11:20:29 -07:00
ntp_internal.h
ntp.c
posix-clock.c
posix-cpu-timers.c
posix-stubs.c
posix-timers.c	posix-timers: Plug potential memory leak in do_timer_create()	2025-11-14 16:58:31 +01:00
posix-timers.h
sched_clock.c	time/sched_clock: Export symbol for sched_clock register function	2025-09-23 10:52:31 +02:00
sleep_timeout.c
test_udelay.c
tick-broadcast-hrtimer.c
tick-broadcast.c
tick-common.c	tick: Do not set device to detached state in tick_shutdown()	2025-09-09 13:39:00 +02:00
tick-internal.h	tick: Do not set device to detached state in tick_shutdown()	2025-09-09 13:39:00 +02:00
tick-legacy.c
tick-oneshot.c
tick-sched.c	tick/sched: Fix bogus condition in report_idle_softirq()	2025-11-19 19:30:45 +01:00
tick-sched.h
time_test.c
time.c	time: export timespec64_add_safe() symbol	2025-09-03 16:51:08 -07:00
timeconst.bc
timeconv.c
timecounter.c
timekeeping_debug.c
timekeeping_internal.h
timekeeping.c	timekeeping: Fix error code in tk_aux_sysfs_init()	2025-11-25 17:52:24 +01:00
timekeeping.h
timer_list.c	hrtimer: Remove hrtimer_clock_base:: Get_time	2025-09-09 12:27:18 +02:00
timer_migration.c	timers/migration: Fix imbalanced NUMA trees	2025-12-18 14:02:47 +01:00
timer_migration.h
timer.c	timers: Fix NULL function pointer race in timer_shutdown_sync()	2025-11-22 22:55:26 +01:00
vsyscall.c	vdso/vsyscall: Avoid slow division loop in auxiliary clock update	2025-09-03 11:55:11 +02:00