From: Nicholas Piggin <npiggin@xxxxxxxxx>
CPU unplug first calls __cpu_disable(), and that's where powerpc calls
cleanup_cpu_mmu_context(), which clears this CPU from mm_cpumask() of all
mms in the system.
However this CPU may still be using a lazy tlb mm, and its mm_cpumask bit
will be cleared from it. The CPU does not switch away from the lazy tlb
mm until arch_cpu_idle_dead() calls idle_task_exit().
If that user mm exits in this window, it will not be subject to the lazy
tlb mm shootdown and may be freed while in use as a lazy mm by the CPU
that is being unplugged.
cleanup_cpu_mmu_context() could be moved later, but it looks better to
move the lazy tlb mm switching earlier. The problem with doing the lazy
mm switching in idle_task_exit() is explained in commit bf2c59fce4074
("sched/core: Fix illegal RCU from offline CPUs"), which added a wart to
switch away from the mm but leave it set in active_mm to be cleaned up
later.
So instead, switch away from the lazy tlb mm at sched_cpu_wait_empty(),
which is the last hotplug state before teardown (CPUHP_AP_SCHED_WAIT_EMPTY).
This CPU will never switch to a user thread from this point, so it has no
chance to pick up a new lazy tlb mm. This removes the lazy tlb mm handling
wart in CPU unplug.
With this, idle_task_exit() is not needed anymore and can be cleaned up.
This leaves the prototype alone, to be cleaned after this change.
herton: took the suggestions from https://lore.kernel.org/all/87jzvyprsw.ffs@tglx/
and made adjustments on the initial patch proposed by Nicholas.
Link: https://lkml.kernel.org/r/20230524060455.147699-1-npiggin@xxxxxxxxx
Link: https://lore.kernel.org/all/20230525205253.E2FAEC433EF@xxxxxxxxxxxxxxx/
Fixes: 2655421ae69fa ("lazy tlb: shoot lazies, non-refcounting lazy tlb mm reference handling scheme")
Signed-off-by: Nicholas Piggin <npiggin@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Suggested-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Signed-off-by: Herton R. Krzesinski <herton@xxxxxxxxxx>
---
include/linux/sched/hotplug.h | 4 ----
kernel/cpu.c | 11 ++++++-----
kernel/sched/core.c | 26 +++++++++++++++++++-------
3 files changed, 25 insertions(+), 16 deletions(-)
Herton: I contacted Nicholas by email, he was ok with me going ahead
and posting this, I saw the original patch was stalled/didn't went forward. Thus
I'm posting this but keeping his From/authorship, since he is original
author of the patch, so we can have this moving forward. I have a report
and also reproduced the warning similar to the one reported at
https://github.com/linuxppc/issues/issues/469 - which can be triggered
doing cpu offline/online loop with CONFIG_DEBUG_VM enabled. This patch fixes
the problem. I updated the changelog/patch based on the suggestions given and to
the best of my knowledge/investigation on this issue, thorough review is
appreciated.
If this is ok then I can submit a followup for this to cleanup idle_task_exit().
diff --git a/include/linux/sched/hotplug.h b/include/linux/sched/hotplug.h
index 412cdaba33eb..17e04859b9a4 100644
--- a/include/linux/sched/hotplug.h
+++ b/include/linux/sched/hotplug.h
@@ -18,10 +18,6 @@ extern int sched_cpu_dying(unsigned int cpu);
# define sched_cpu_dying NULL
#endif
-#ifdef CONFIG_HOTPLUG_CPU
-extern void idle_task_exit(void);
-#else
static inline void idle_task_exit(void) {}
-#endif
#endif /* _LINUX_SCHED_HOTPLUG_H */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index d293d52a3e00..fb4f46885cb2 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -904,13 +904,14 @@ static int finish_cpu(unsigned int cpu)
struct task_struct *idle = idle_thread_get(cpu);
struct mm_struct *mm = idle->active_mm;
- /*
- * idle_task_exit() will have switched to &init_mm, now
- * clean up any remaining active_mm state.
+ /*
+ * sched_force_init_mm() ensured the use of &init_mm,
+ * drop that refcount now that the CPU has stopped.
*/
- if (mm != &init_mm)
- idle->active_mm = &init_mm;
+ WARN_ON(mm != &init_mm);
+ idle->active_mm = NULL;
mmdrop_lazy_tlb(mm);
+
return 0;
}
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index dbfb5717d6af..8a51acd6d650 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -7826,19 +7826,26 @@ void sched_setnuma(struct task_struct *p, int nid)
#ifdef CONFIG_HOTPLUG_CPU
/*
- * Ensure that the idle task is using init_mm right before its CPU goes
- * offline.
+ * Invoked on the outgoing CPU in context of the CPU hotplug thread
+ * after ensuring that there are no user space tasks left on the CPU.
+ *
+ * If there is a lazy mm in use on the hotplug thread, drop it and
+ * switch to init_mm.
+ *
+ * The reference count on init_mm is dropped in finish_cpu().
*/
-void idle_task_exit(void)
+static void sched_force_init_mm(void)
{
struct mm_struct *mm = current->active_mm;
- BUG_ON(cpu_online(smp_processor_id()));
- BUG_ON(current != this_rq()->idle);
-
if (mm != &init_mm) {
- switch_mm(mm, &init_mm, current);
+ mmgrab_lazy_tlb(&init_mm);
+ local_irq_disable();
+ current->active_mm = &init_mm;
+ switch_mm_irqs_off(mm, &init_mm, current);
+ local_irq_enable();
finish_arch_post_lock_switch();
+ mmdrop_lazy_tlb(mm);
}
/* finish_cpu(), as ran on the BP, will clean up the active_mm state */
@@ -7972,6 +7979,10 @@ static void balance_hotplug_wait(void)
#else
+static void sched_force_init_mm(void)
+{
+}
+
static inline void balance_push(struct rq *rq)
{
}
@@ -8240,6 +8251,7 @@ int sched_cpu_starting(unsigned int cpu)
int sched_cpu_wait_empty(unsigned int cpu)
{
balance_hotplug_wait();
+ sched_force_init_mm();
return 0;
}
--
2.46.2
