Hi,
The updated queue of moving from a push remote at enqueue time to a pull at
expiry time model now also contains changes regarding the timer base idle
marking.
The queue is splitted into three parts:
- Patches 1 - 7: Cleanups and minor fixes
- Patches 8 - 10: timer base idle marking rework with two preparatory
changes. See the section below for more details.
- Patches 11 - 25: Updated timer pull model on top of timer idle rework
The queue is available here:
git://git.kernel.org/pub/scm/linux/kernel/git/tglx/devel timers/pushpull
Move marking timer bases as idle into tick_nohz_stop_tick()
-----------------------------------------------------------
The idle marking of timer bases is done in get_next_timer_interrupt()
whenever possible. The timer bases are idle, even if the tick will not be
stopped. This lead to an IPI when a new first timer is enqueued remote. To
prevent this, setting timer_base->in_idle flag is postponed to
tick_nohz_stop_tick().
Timer pull model
----------------
Placing timers at enqueue time on a target CPU based on dubious heuristics
does not make any sense:
1) Most timer wheel timers are canceled or rearmed before they expire.
2) The heuristics to predict which CPU will be busy when the timer expires
are wrong by definition.
So placing the timers at enqueue wastes precious cycles.
The proper solution to this problem is to always queue the timers on the
local CPU and allow the non pinned timers to be pulled onto a busy CPU at
expiry time.
Therefore split the timer storage into local pinned and global timers:
Local pinned timers are always expired on the CPU on which they have been
queued. Global timers can be expired on any CPU.
As long as a CPU is busy it expires both local and global timers. When a
CPU goes idle it arms for the first expiring local timer. If the first
expiring pinned (local) timer is before the first expiring movable timer,
then no action is required because the CPU will wake up before the first
movable timer expires. If the first expiring movable timer is before the
first expiring pinned (local) timer, then this timer is queued into a idle
timerqueue and eventually expired by some other active CPU.
To avoid global locking the timerqueues are implemented as a hierarchy. The
lowest level of the hierarchy holds the CPUs. The CPUs are associated to
groups of 8, which are separated per node. If more than one CPU group
exist, then a second level in the hierarchy collects the groups. Depending
on the size of the system more than 2 levels are required. Each group has a
"migrator" which checks the timerqueue during the tick for remote timers to
be expired.
If the last CPU in a group goes idle it reports the first expiring event in
the group up to the next group(s) in the hierarchy. If the last CPU goes
idle it arms its timer for the first system wide expiring timer to ensure
that no timer event is missed.
Testing
~~~~~~~
Enqueue
^^^^^^^
The impact of wasting cycles during enqueue by using the heuristic in
contrast to always queuing the timer on the local CPU was measured with a
micro benchmark. Therefore a timer is enqueued and dequeued in a loop with
1000 repetitions on a isolated CPU. The time the loop takes is measured. A
quarter of the remaining CPUs was kept busy. This measurement was repeated
several times. With the patch queue the average duration was reduced by
approximately 25%.
145ns plain v6
109ns v6 with patch queue
Furthermore the impact of residence in deep idle states of an idle system
was investigated. The patch queue doesn't downgrade this behavior.
dbench test
^^^^^^^^^^^
A dbench test starting X pairs of client servers are used to create load on
the system. The measurable value is the throughput. The tests were executed
on a zen3 machine. The base is the tip tree branch timers/core which is
based on a v6.6-rc1.
governor menu
X pairs timers/core pull-model impact
----------------------------------------------
1 353.19 (0.19) 353.45 (0.30) 0.07%
2 700.10 (0.96) 687.00 (0.20) -1.87%
4 1329.37 (0.63) 1282.91 (0.64) -3.49%
8 2561.16 (1.28) 2493.56 (1.76) -2.64%
16 4959.96 (0.80) 4914.59 (0.64) -0.91%
32 9741.92 (3.44) 8979.83 (1.13) -7.82%
64 16535.40 (2.84) 16388.47 (4.02) -0.89%
128 22136.83 (2.42) 23174.50 (1.43) 4.69%
256 39256.77 (4.48) 38994.00 (0.39) -0.67%
512 36799.03 (1.83) 38091.10 (0.63) 3.51%
1024 32903.03 (0.86) 35370.70 (0.89) 7.50%
governor teo
X pairs timers/core pull-model impact
----------------------------------------------
1 350.83 (1.27) 352.45 (0.96) 0.46%
2 699.52 (0.85) 690.10 (0.54) -1.35%
4 1339.53 (1.99) 1294.71 (2.71) -3.35%
8 2574.10 (0.76) 2495.46 (1.97) -3.06%
16 4898.50 (1.74) 4783.06 (1.64) -2.36%
32 9115.50 (4.63) 9037.83 (1.58) -0.85%
64 16663.90 (3.80) 16042.00 (1.72) -3.73%
128 25044.93 (1.11) 23250.03 (1.08) -7.17%
256 38059.53 (1.70) 39658.57 (2.98) 4.20%
512 36369.30 (0.39) 38890.13 (0.36) 6.93%
1024 33956.83 (1.14) 35514.83 (0.29) 4.59%
Ping Pong Oberservation
^^^^^^^^^^^^^^^^^^^^^^^
During testing on a mostly idle machine a ping pong game could be observed:
a process_timeout timer is expired remotely on a non idle CPU. Then the CPU
where the schedule_timeout() was executed to enqueue the timer comes out of
idle and restarts the timer using schedule_timeout() and goes back to idle
again. This is due to the fair scheduler which tries to keep the task on
the CPU which it previously executed on.
Possible Next Steps
~~~~~~~~~~~~~~~~~~~
Simple deferrable timers are no longer required as they can be converted to
global timers. If a CPU goes idle, a formerly deferrable timer will not
prevent the CPU to sleep as long as possible. Only the last migrator CPU
has to take care of them. Deferrable timers with timer pinned flags needs
to be expired on the specified CPU but must not prevent CPU from going
idle. They require their own timer base which is never taken into account
when calculating the next expiry time. This conversation and required
cleanup will be done in a follow up series.
v7..v8: https://lore.kernel.org/r/[email protected]
- Address review feedback
- Move marking timer base idle into tick_nohz_stop_tick()
- Look ahead function to determine possible sleep lenght
v6..v7:
- Address review feedback of Frederic and bigeasy
- Change lock, unlock fetch next timer interrupt logic after remote expiry
- Move timer_expire_remote() into tick-internal.h
- Add documentation section about "Required event and timerqueue update
after remote expiry"
- Fix fallout of kernel test robot
v5..v6:
- Address review of Frederic Weisbecker and Peter Zijlstra (spelling,
locking, race in tmigr_handle_remote_cpu())
- unconditionally set TIMER_PINNED flag in add_timer_on(); introduce
add_timer() variants which set/unset TIMER_PINNED flag; drop fixing
add_timer_on() call sites, as TIMER_PINNED flag is set implicitly;
Fixing workqueue to use add_timer_global() instead of simply
add_timer() for unbound work.
- Drop support for siblings to end up in the same level 0 group (could be
added again in a better way as an improvement later on)
- Do not send IPI for new first deferrable timers
v4..v5:
- address review feedback of Frederic Weisbecker
- fix issue with group timer update after remote expiry
v3..v4:
- address review feedback of Frederic Weisbecker
- address kernel test robot fallout
- Move patch 16 "add_timer_on(): Make sure callers have TIMER_PINNED
flag" at the begin of the queue to prevent timers to end up in global
timer base when they were queued using add_timer_on()
- Fix some comments and typos
v2..v3: https://lore.kernel.org/r/[email protected]/
- Minimize usage of locks by storing data using atomic_cmpxchg() for
migrator information and information about active cpus.
Thanks,
Anna-Maria
Anna-Maria Behnsen (22):
tick/sched: Cleanup confusing variables
tick-sched: Warn when next tick seems to be in the past
timer: Do not IPI for deferrable timers
timer: Move store of next event into __next_timer_interrupt()
timers: Clarify check in forward_timer_base()
timers: Split out forward timer base functionality
timers: Use already existing function for forwarding timer base
timer: Split out get next timer functionality
timers: Move marking timer bases idle into tick_nohz_stop_tick()
timers: Introduce add_timer() variants which modify timer flags
workqueue: Use global variant for add_timer()
timer: add_timer_on(): Make sure TIMER_PINNED flag is set
timers: Ease code in run_local_timers()
timer: Split next timer interrupt logic
timer: Keep the pinned timers separate from the others
timer: Retrieve next expiry of pinned/non-pinned timers separately
timer: Split out "get next timer interrupt" functionality
timer: Add get next timer interrupt functionality for remote CPUs
timer: Check if timers base is handled already
timer: Implement the hierarchical pull model
timer_migration: Add tracepoints
timer: Always queue timers on the local CPU
Richard Cochran (linutronix GmbH) (2):
timer: Restructure internal locking
tick/sched: Split out jiffies update helper function
Thomas Gleixner (1):
timer: Rework idle logic
include/linux/cpuhotplug.h | 1 +
include/linux/timer.h | 16 +-
include/trace/events/timer_migration.h | 283 ++++
kernel/time/Makefile | 3 +
kernel/time/tick-internal.h | 13 +
kernel/time/tick-sched.c | 69 +-
kernel/time/timer.c | 514 ++++++--
kernel/time/timer_migration.c | 1636 ++++++++++++++++++++++++
kernel/time/timer_migration.h | 144 +++
kernel/workqueue.c | 2 +-
10 files changed, 2549 insertions(+), 132 deletions(-)
create mode 100644 include/trace/events/timer_migration.h
create mode 100644 kernel/time/timer_migration.c
create mode 100644 kernel/time/timer_migration.h
--
2.39.2
Deferrable timers do not prevent CPU from going idle and are not taken into
account on idle path. Sending an IPI to a remote CPU when a new first
deferrable timer was enqueued will wake up the remote CPU and but nothing
will be done regarding the deferrable timers.
Drop IPI completely when a new first deferrable timer was enqueued.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
v8: Update comment
v6: new patch
---
kernel/time/timer.c | 15 ++++++---------
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 63a8ce7177dd..7d06351aff54 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -571,18 +571,15 @@ static int calc_wheel_index(unsigned long expires, unsigned long clk,
static void
trigger_dyntick_cpu(struct timer_base *base, struct timer_list *timer)
{
- if (!is_timers_nohz_active())
- return;
-
/*
- * TODO: This wants some optimizing similar to the code below, but we
- * will do that when we switch from push to pull for deferrable timers.
+ * Deferrable timers do not prevent CPU from entering dynticks and are
+ * not taken into account on idle/nohz_full path. An IPI when a new
+ * deferrable timer is enqueued will wake up the remote CPU but nothing
+ * will be done with the deferrable timer base. Therefore skip remote
+ * IPI for deferrable timers completely.
*/
- if (timer->flags & TIMER_DEFERRABLE) {
- if (tick_nohz_full_cpu(base->cpu))
- wake_up_nohz_cpu(base->cpu);
+ if (!is_timers_nohz_active() || timer->flags & TIMER_DEFERRABLE)
return;
- }
/*
* We might have to IPI the remote CPU if the base is idle and the
--
2.39.2
Logic for getting next timer interrupt (no matter of recalculated or
already stored in base->next_expiry) is split into a separate function
"next_timer_interrupt()" to make it available for new call sites.
No functional change.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
kernel/time/timer.c | 12 +++++++++---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 1aafa0edc74c..7468a805d302 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1951,14 +1951,20 @@ static u64 cmp_next_hrtimer_event(u64 basem, u64 expires)
return DIV_ROUND_UP_ULL(nextevt, TICK_NSEC) * TICK_NSEC;
}
+static unsigned long next_timer_interrupt(struct timer_base *base)
+{
+ if (base->next_expiry_recalc)
+ next_expiry_recalc(base);
+
+ return base->next_expiry;
+}
+
static inline unsigned long __get_next_timer_interrupt(unsigned long basej,
struct timer_base *base)
{
unsigned long nextevt;
- if (base->next_expiry_recalc)
- next_expiry_recalc(base);
- nextevt = base->next_expiry;
+ nextevt = next_timer_interrupt(base);
if (base->timers_pending) {
/* If we missed a tick already, force 0 delta */
--
2.39.2
From: Thomas Gleixner <[email protected]>
To improve readability of the code, split base->idle calculation and
expires calculation into separate parts.
Thereby the following subtle change happens if the next event is just one
jiffy ahead and the tick was already stopped: Originally base->is_idle
remains true in this situation. Now base->is_idle turns to false. This may
spare an IPI if a timer is enqueued remotely to an idle CPU that is going
to tick on the next jiffy.
Signed-off-by: Thomas Gleixner <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
v4: Change condition to force 0 delta and update commit message (Frederic)
---
kernel/time/timer.c | 29 ++++++++++++++---------------
1 file changed, 14 insertions(+), 15 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index dc58c479d35a..18f8aac9b19a 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1943,21 +1943,20 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
*/
__forward_timer_base(base, basej);
- if (time_before_eq(nextevt, basej)) {
- expires = basem;
- base->is_idle = false;
- } else {
- if (base->timers_pending)
- expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
- /*
- * If we expect to sleep more than a tick, mark the base idle.
- * Also the tick is stopped so any added timer must forward
- * the base clk itself to keep granularity small. This idle
- * logic is only maintained for the BASE_STD base, deferrable
- * timers may still see large granularity skew (by design).
- */
- if ((expires - basem) > TICK_NSEC)
- base->is_idle = true;
+ /*
+ * Base is idle if the next event is more than a tick away. Also
+ * the tick is stopped so any added timer must forward the base clk
+ * itself to keep granularity small. This idle logic is only
+ * maintained for the BASE_STD base, deferrable timers may still
+ * see large granularity skew (by design).
+ */
+ base->is_idle = time_after(nextevt, basej + 1);
+
+ if (base->timers_pending) {
+ /* If we missed a tick already, force 0 delta */
+ if (time_before(nextevt, basej))
+ nextevt = basej;
+ expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
}
raw_spin_unlock(&base->lock);
--
2.39.2
The implementation of the NOHZ pull at expiry model will change the timer
bases per CPU. Timers, that have to expire on a specific CPU, require the
TIMER_PINNED flag. If the CPU doesn't matter, the TIMER_PINNED flag must be
dropped. This is required for call sites which use the timer alternately as
pinned and not pinned timer like workqueues do.
Therefore use add_timer_global() to make sure TIMER_PINNED flag is dropped.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Lai Jiangshan <[email protected]>
---
v6:
- New patch: As v6 provides unconditially setting TIMER_PINNED flag in
add_timer_on() workqueue requires new add_timer_global() variant.
---
kernel/workqueue.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index c85825e17df8..36474a2c4b6f 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -1958,7 +1958,7 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq,
if (unlikely(cpu != WORK_CPU_UNBOUND))
add_timer_on(timer, cpu);
else
- add_timer(timer);
+ add_timer_global(timer);
}
/**
--
2.39.2
When adding a timer to the timer wheel using add_timer_on(), it is an
implicitly pinned timer. With the timer pull at expiry time model in place,
TIMER_PINNED flag is required to make sure timers end up in proper base.
Add TIMER_PINNED flag unconditionally when add_timer_on() is executed.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
kernel/time/timer.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 3172b8094097..853a8ee48ffc 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1284,7 +1284,10 @@ EXPORT_SYMBOL(add_timer_global);
* @timer: The timer to be started
* @cpu: The CPU to start it on
*
- * Same as add_timer() except that it starts the timer on the given CPU.
+ * Same as add_timer() except that it starts the timer on the given CPU and
+ * the TIMER_PINNED flag is set. When timer shouldn't be a pinned timer in
+ * the next round, add_timer_global() should be used instead as it unsets
+ * the TIMER_PINNED flag.
*
* See add_timer() for further details.
*/
@@ -1298,6 +1301,9 @@ void add_timer_on(struct timer_list *timer, int cpu)
if (WARN_ON_ONCE(timer_pending(timer)))
return;
+ /* Make sure timer flags have TIMER_PINNED flag set */
+ timer->flags |= TIMER_PINNED;
+
new_base = get_timer_cpu_base(timer->flags, cpu);
/*
--
2.39.2
The timer base is marked idle when get_next_timer_interrupt() is
executed. But the decision whether the tick will be stopped and whether the
system is able to go idle is done later. When the timer bases is marked
idle and a new first timer is enqueued remote an IPI is raised. Even if it
is not required because the tick is not stopped and the timer base is
evaluated again at the next tick.
To prevent this, the timer base is marked idle in tick_nohz_stop_tick() and
get_next_timer_interrupt() is streamlined by only looking for the next
timer interrupt. All other work is postponed to timer_set_idle() which is
called by tick_nohz_stop_tick().
While at it a whitespace damage is fixed as well.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
kernel/time/tick-internal.h | 1 +
kernel/time/tick-sched.c | 38 ++++++++++++++++++++++++-----------
kernel/time/timer.c | 40 +++++++++++++++++++++++++++++++++----
3 files changed, 63 insertions(+), 16 deletions(-)
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 649f2b48e8f0..b035606a6f5e 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -164,6 +164,7 @@ static inline void timers_update_nohz(void) { }
DECLARE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases);
extern u64 get_next_timer_interrupt(unsigned long basej, u64 basem);
+u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle);
void timer_clear_idle(void);
#define CLOCK_SET_WALL \
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index b3cf535881a4..7e1fdbc6d5f0 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -846,11 +846,6 @@ static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
delta = next_tick - basemono;
if (delta <= (u64)TICK_NSEC) {
- /*
- * Tell the timer code that the base is not idle, i.e. undo
- * the effect of get_next_timer_interrupt():
- */
- timer_clear_idle();
/*
* We've not stopped the tick yet, and there's a timer in the
* next period, so no point in stopping it either, bail.
@@ -886,12 +881,35 @@ static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
{
struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
+ unsigned long basejiff = ts->last_jiffies;
u64 basemono = ts->timer_expires_base;
- u64 expires = ts->timer_expires;
+ bool timer_idle = ts->tick_stopped;
+ u64 expires;
/* Make sure we won't be trying to stop it twice in a row. */
ts->timer_expires_base = 0;
+ /*
+ * Now the tick should be stopped definitely - so timer base needs to be
+ * marked idle as well to not miss a newly queued timer.
+ */
+ expires = timer_set_idle(basejiff, basemono, &timer_idle);
+ if (!timer_idle) {
+ /*
+ * Do not clear tick_stopped here when it was already set - it will
+ * be retained on next idle iteration when tick expired earlier
+ * than expected.
+ */
+ expires = basemono + TICK_NSEC;
+
+ /* Undo the effect of timer_set_idle() */
+ timer_clear_idle();
+ } else if (expires < ts->timer_expires) {
+ ts->timer_expires = expires;
+ } else {
+ expires = ts->timer_expires;
+ }
+
/*
* If this CPU is the one which updates jiffies, then give up
* the assignment and let it be taken by the CPU which runs
@@ -926,7 +944,7 @@ static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
* first call we save the current tick time, so we can restart
* the scheduler tick in nohz_restart_sched_tick.
*/
- if (!ts->tick_stopped) {
+ if (!ts->tick_stopped && timer_idle) {
calc_load_nohz_start();
quiet_vmstat();
@@ -989,7 +1007,7 @@ static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now)
/*
* Cancel the scheduled timer and restore the tick
*/
- ts->tick_stopped = 0;
+ ts->tick_stopped = 0;
tick_nohz_restart(ts, now);
}
@@ -1145,10 +1163,6 @@ void tick_nohz_idle_stop_tick(void)
void tick_nohz_idle_retain_tick(void)
{
tick_nohz_retain_tick(this_cpu_ptr(&tick_cpu_sched));
- /*
- * Undo the effect of get_next_timer_interrupt() called from
- * tick_nohz_next_event().
- */
timer_clear_idle();
}
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index f443aa807fbc..8518f7aa7319 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1950,6 +1950,40 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
if (cpu_is_offline(smp_processor_id()))
return expires;
+ raw_spin_lock(&base->lock);
+ nextevt = __get_next_timer_interrupt(basej, base);
+ raw_spin_unlock(&base->lock);
+
+ expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
+
+ return cmp_next_hrtimer_event(basem, expires);
+}
+
+/**
+ * timer_set_idle - Set the idle state of the timer bases (if possible)
+ * @basej: base time jiffies
+ * @basem: base time clock monotonic
+ * @idle: pointer to store the value of timer_base->in_idle
+ *
+ * Returns the next timer expiry.
+ *
+ * hrtimers are not taken into account once more, as they already have been
+ * taken into account when asking for the next timer expiry.
+ */
+u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
+{
+ struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
+ unsigned long nextevt;
+
+ /*
+ * Pretend that there is no timer pending if the cpu is offline.
+ * Possible pending timers will be migrated later to an active cpu.
+ */
+ if (cpu_is_offline(smp_processor_id())) {
+ *idle = true;
+ return KTIME_MAX;
+ }
+
raw_spin_lock(&base->lock);
nextevt = __get_next_timer_interrupt(basej, base);
@@ -1966,13 +2000,11 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
* maintained for the BASE_STD base, deferrable timers may still
* see large granularity skew (by design).
*/
- base->is_idle = time_after(nextevt, basej + 1);
+ base->is_idle = *idle = time_after(nextevt, basej + 1);
raw_spin_unlock(&base->lock);
- expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
-
- return cmp_next_hrtimer_event(basem, expires);
+ return basem + (u64)(nextevt - basej) * TICK_NSEC;
}
/**
--
2.39.2
For the conversion of the NOHZ timer placement to a pull at expiry time
model it's required to have separate expiry times for the pinned and the
non-pinned (movable) timers. Therefore struct timer_events is introduced.
No functional change
Originally-by: Richard Cochran (linutronix GmbH) <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
kernel/time/timer.c | 48 ++++++++++++++++++++++++++++++++++-----------
1 file changed, 37 insertions(+), 11 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 1cd214c2f3f1..4230fc6fa1ed 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -221,6 +221,11 @@ struct timer_base {
static DEFINE_PER_CPU(struct timer_base, timer_bases[NR_BASES]);
+struct timer_events {
+ u64 local;
+ u64 global;
+};
+
#ifdef CONFIG_NO_HZ_COMMON
static DEFINE_STATIC_KEY_FALSE(timers_nohz_active);
@@ -1971,9 +1976,10 @@ static unsigned long next_timer_interrupt(struct timer_base *base)
return base->next_expiry;
}
-static inline unsigned long __get_next_timer_interrupt(unsigned long basej,
+static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64 basem,
struct timer_base *base_local,
- struct timer_base *base_global)
+ struct timer_base *base_global,
+ struct timer_events *tevt)
{
unsigned long nextevt, nextevt_local, nextevt_global;
bool local_first;
@@ -1996,12 +2002,32 @@ static inline unsigned long __get_next_timer_interrupt(unsigned long basej,
nextevt = local_first ? nextevt_local : nextevt_global;
- if (base_local->timers_pending || base_global->timers_pending) {
+ /*
+ * If the @nextevt is at max. one tick away, use @nextevt and store
+ * it in the local expiry value. The next global event is irrelevant in
+ * this case and can be left as KTIME_MAX.
+ */
+ if (time_before_eq(nextevt, basej + 1)) {
/* If we missed a tick already, force 0 delta */
if (time_before(nextevt, basej))
nextevt = basej;
+ tevt->local = basem + (u64)(nextevt - basej) * TICK_NSEC;
+ return tevt->local;
}
+ /*
+ * Update tevt->* values:
+ *
+ * If the local queue expires first, then the global event can
+ * be ignored. If the global queue is empty, nothing to do
+ * either.
+ */
+ if (!local_first && base_global->timers_pending)
+ tevt->global = basem + (u64)(nextevt_global - basej) * TICK_NSEC;
+
+ if (base_local->timers_pending)
+ tevt->local = basem + (u64)(nextevt_local - basej) * TICK_NSEC;
+
return nextevt;
}
@@ -2015,16 +2041,15 @@ static inline unsigned long __get_next_timer_interrupt(unsigned long basej,
*/
u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
{
+ struct timer_events tevt = { .local = KTIME_MAX, .global = KTIME_MAX };
struct timer_base *base_local, *base_global;
- u64 expires = KTIME_MAX;
- unsigned long nextevt;
/*
* Pretend that there is no timer pending if the cpu is offline.
* Possible pending timers will be migrated later to an active cpu.
*/
if (cpu_is_offline(smp_processor_id()))
- return expires;
+ return KTIME_MAX;
base_local = this_cpu_ptr(&timer_bases[BASE_LOCAL]);
base_global = this_cpu_ptr(&timer_bases[BASE_GLOBAL]);
@@ -2032,14 +2057,14 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
raw_spin_lock(&base_local->lock);
raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
- nextevt = __get_next_timer_interrupt(basej, base_local, base_global);
+ __get_next_timer_interrupt(basej, basem, base_local, base_global, &tevt);
raw_spin_unlock(&base_global->lock);
raw_spin_unlock(&base_local->lock);
- expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
+ tevt.local = min_t(u64, tevt.local, tevt.global);
- return cmp_next_hrtimer_event(basem, expires);
+ return cmp_next_hrtimer_event(basem, tevt.local);
}
/**
@@ -2055,6 +2080,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
*/
u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
{
+ struct timer_events tevt = { .local = KTIME_MAX, .global = KTIME_MAX };
struct timer_base *base_local, *base_global;
unsigned long nextevt;
@@ -2073,7 +2099,7 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
raw_spin_lock(&base_local->lock);
raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
- nextevt = __get_next_timer_interrupt(basej, base_local, base_global);
+ nextevt = __get_next_timer_interrupt(basej, basem, base_local, base_global, &tevt);
/*
* We have a fresh next event. Check whether we can forward the
@@ -2097,7 +2123,7 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
raw_spin_unlock(&base_global->lock);
raw_spin_unlock(&base_local->lock);
- return basem + (u64)(nextevt - basej) * TICK_NSEC;
+ return min_t(u64, tevt.local, tevt.global);
}
/**
--
2.39.2
From: "Richard Cochran (linutronix GmbH)" <[email protected]>
The logic to get the time of the last jiffies update will be needed by
the timer pull model as well.
Move the code into a global function in anticipation of the new caller.
No functional change.
Signed-off-by: Richard Cochran (linutronix GmbH) <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
kernel/time/tick-internal.h | 1 +
kernel/time/tick-sched.c | 18 +++++++++++++++---
2 files changed, 16 insertions(+), 3 deletions(-)
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 206010ae2a53..7dbf5ab43d06 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -158,6 +158,7 @@ static inline void tick_nohz_init(void) { }
#ifdef CONFIG_NO_HZ_COMMON
extern unsigned long tick_nohz_active;
extern void timers_update_nohz(void);
+extern u64 get_jiffies_update(unsigned long *basej);
# ifdef CONFIG_SMP
extern struct static_key_false timers_migration_enabled;
extern void fetch_next_timer_interrupt_remote(unsigned long basej, u64 basem,
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 7e1fdbc6d5f0..169886a87cc1 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -798,18 +798,30 @@ static inline bool local_timer_softirq_pending(void)
return local_softirq_pending() & BIT(TIMER_SOFTIRQ);
}
-static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
+/*
+ * Read jiffies and the time when jiffies were updated last
+ */
+u64 get_jiffies_update(unsigned long *basej)
{
- u64 basemono, next_tick, delta, expires;
unsigned long basejiff;
unsigned int seq;
+ u64 basemono;
- /* Read jiffies and the time when jiffies were updated last */
do {
seq = read_seqcount_begin(&jiffies_seq);
basemono = last_jiffies_update;
basejiff = jiffies;
} while (read_seqcount_retry(&jiffies_seq, seq));
+ *basej = basejiff;
+ return basemono;
+}
+
+static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
+{
+ u64 basemono, next_tick, delta, expires;
+ unsigned long basejiff;
+
+ basemono = get_jiffies_update(&basejiff);
ts->last_jiffies = basejiff;
ts->timer_expires_base = basemono;
--
2.39.2
Placing timers at enqueue time on a target CPU based on dubious heuristics
does not make any sense:
1) Most timer wheel timers are canceled or rearmed before they expire.
2) The heuristics to predict which CPU will be busy when the timer expires
are wrong by definition.
So placing the timers at enqueue wastes precious cycles.
The proper solution to this problem is to always queue the timers on the
local CPU and allow the non pinned timers to be pulled onto a busy CPU at
expiry time.
Therefore split the timer storage into local pinned and global timers:
Local pinned timers are always expired on the CPU on which they have been
queued. Global timers can be expired on any CPU.
As long as a CPU is busy it expires both local and global timers. When a
CPU goes idle it arms for the first expiring local timer. If the first
expiring pinned (local) timer is before the first expiring movable timer,
then no action is required because the CPU will wake up before the first
movable timer expires. If the first expiring movable timer is before the
first expiring pinned (local) timer, then this timer is queued into a idle
timerqueue and eventually expired by some other active CPU.
To avoid global locking the timerqueues are implemented as a hierarchy. The
lowest level of the hierarchy holds the CPUs. The CPUs are associated to
groups of 8, which are separated per node. If more than one CPU group
exist, then a second level in the hierarchy collects the groups. Depending
on the size of the system more than 2 levels are required. Each group has a
"migrator" which checks the timerqueue during the tick for remote expirable
timers.
If the last CPU in a group goes idle it reports the first expiring event in
the group up to the next group(s) in the hierarchy. If the last CPU goes
idle it arms its timer for the first system wide expiring timer to ensure
that no timer event is missed.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
v8:
- Review of Frederic:
- Fix hotplug race (introduction of wakeup_recalc)
- Make wakeup and wakeup_recalc logic consistent all over the place
- Fix child/group state race and read it with locks held
- Add more clarifying comments
- Fix grammar all over the place
- change integers which act as boolean value into bool
- rewrite condition in tmigr_check_migrator() without negation
- Improve update events logic with a check of the first event
- Implement a quick forecast which is called when
get_next_timer_interrupt() is executed.
v7:
- Review remarks of Frederic and bigeasy:
- change logic in tmigr_handle_remote_cpu()
- s/kzalloc/kcalloc
- move timer_expire_remote() into NO_HZ_COMMON && SMP config section
- drop DBG_BUG_ON() makro and use only WARN_ON_ONCE()
- remove leftovers from sibling logic during setup
- Move timer_expire_remote() into tick-internal.h
- Add documentation section about "Required event and timerqueue update
after remote expiry"
- Fix fallout of kernel test robot
v6:
- Fix typos
- Review remarks of Peter Zijlstra (locking, struct member cleanup, use
atomic_try_cmpxchg(), update struct member descriptions)
- Fix race in tmigr_handle_remote_cpu() (Frederic Weisbecker)
v5:
- Review remarks of Frederic
- Return nextevt when CPU is marked offline in timer migration hierarchy
instead of KTIME_MAX
- Fix update of group events issue, after remote expiring
v4:
- Fold typo fix in comment into proper patch "timer: Split out "get next
timer interrupt" functionality"
- Update wrong comment for tmigr_state union definition
- Fix fallout of kernel test robot
---
include/linux/cpuhotplug.h | 1 +
kernel/time/Makefile | 3 +
kernel/time/tick-internal.h | 1 +
kernel/time/timer.c | 81 +-
kernel/time/timer_migration.c | 1612 +++++++++++++++++++++++++++++++++
kernel/time/timer_migration.h | 144 +++
6 files changed, 1836 insertions(+), 6 deletions(-)
create mode 100644 kernel/time/timer_migration.c
create mode 100644 kernel/time/timer_migration.h
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index 06dda85f0424..ed992268f206 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -244,6 +244,7 @@ enum cpuhp_state {
CPUHP_AP_PERF_POWERPC_HV_24x7_ONLINE,
CPUHP_AP_PERF_POWERPC_HV_GPCI_ONLINE,
CPUHP_AP_PERF_CSKY_ONLINE,
+ CPUHP_AP_TMIGR_ONLINE,
CPUHP_AP_WATCHDOG_ONLINE,
CPUHP_AP_WORKQUEUE_ONLINE,
CPUHP_AP_RANDOM_ONLINE,
diff --git a/kernel/time/Makefile b/kernel/time/Makefile
index 7e875e63ff3b..4af2a264a160 100644
--- a/kernel/time/Makefile
+++ b/kernel/time/Makefile
@@ -17,6 +17,9 @@ endif
obj-$(CONFIG_GENERIC_SCHED_CLOCK) += sched_clock.o
obj-$(CONFIG_TICK_ONESHOT) += tick-oneshot.o tick-sched.o
obj-$(CONFIG_LEGACY_TIMER_TICK) += tick-legacy.o
+ifeq ($(CONFIG_SMP),y)
+ obj-$(CONFIG_NO_HZ_COMMON) += timer_migration.o
+endif
obj-$(CONFIG_HAVE_GENERIC_VDSO) += vsyscall.o
obj-$(CONFIG_DEBUG_FS) += timekeeping_debug.o
obj-$(CONFIG_TEST_UDELAY) += test_udelay.o
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 7dbf5ab43d06..fb1568259c21 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -166,6 +166,7 @@ extern void fetch_next_timer_interrupt_remote(unsigned long basej, u64 basem,
unsigned int cpu);
extern void timer_lock_remote_bases(unsigned int cpu);
extern void timer_unlock_remote_bases(unsigned int cpu);
+extern void timer_expire_remote(unsigned int cpu);
# endif
#else /* CONFIG_NO_HZ_COMMON */
static inline void timers_update_nohz(void) { }
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 7f06553e44de..331c59c0789a 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -53,6 +53,7 @@
#include <asm/io.h>
#include "tick-internal.h"
+#include "timer_migration.h"
#define CREATE_TRACE_POINTS
#include <trace/events/timer.h>
@@ -2080,6 +2081,21 @@ void timer_lock_remote_bases(unsigned int cpu)
raw_spin_lock(&base_local->lock);
raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
}
+
+static void __run_timer_base(struct timer_base *base);
+
+/**
+ * timer_expire_remote - expire global timers of cpu
+ * @cpu: Remote CPU
+ *
+ * Expire timers of global base of remote CPU.
+ */
+void timer_expire_remote(unsigned int cpu)
+{
+ struct timer_base *base = per_cpu_ptr(&timer_bases[BASE_GLOBAL], cpu);
+
+ __run_timer_base(base);
+}
# endif /* CONFIG_SMP */
static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64 basem,
@@ -2113,13 +2129,18 @@ static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64
* @basej: base time jiffies
* @basem: base time clock monotonic
*
- * Returns the tick aligned clock monotonic time of the next pending
- * timer or KTIME_MAX if no timer is pending.
+ * Returns the tick aligned clock monotonic time of the next pending timer
+ * or KTIME_MAX if no timer is pending. If timer of global base was queued
+ * into timer migration hierarchy, first global timer is not taken into
+ * account. If it was the last CPU of timer migration hierarchy going idle,
+ * first global event is taken into account.
*/
u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
{
struct timer_events tevt = { .local = KTIME_MAX, .global = KTIME_MAX };
struct timer_base *base_local, *base_global;
+ u64 tmigr_expires = KTIME_MAX;
+ unsigned long nextevt;
/*
* Pretend that there is no timer pending if the cpu is offline.
@@ -2134,12 +2155,19 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
raw_spin_lock(&base_local->lock);
raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
- __get_next_timer_interrupt(basej, basem, base_local, base_global, &tevt);
+ nextevt = __get_next_timer_interrupt(basej, basem, base_local, base_global, &tevt);
+
+ if (time_after(nextevt, basej + 1)) {
+ tmigr_expires = tmigr_quick_check();
+
+ if (tmigr_expires < basem)
+ tmigr_expires = basem;
+ }
raw_spin_unlock(&base_global->lock);
raw_spin_unlock(&base_local->lock);
- tevt.local = min_t(u64, tevt.local, tevt.global);
+ tevt.local = min_t(u64, tevt.local, tmigr_expires);
return cmp_next_hrtimer_event(basem, tevt.local);
}
@@ -2194,13 +2222,47 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
*/
*idle = time_after(nextevt, basej + 1);
+ if (*idle) {
+ u64 next_tmigr;
+
+ /*
+ * Enqueue first global timer into timer migration
+ * hierarchy, afterwards tevt.global is no longer used.
+ */
+ next_tmigr = tmigr_cpu_deactivate(tevt.global);
+
+ /*
+ * If CPU is the last going idle in timer migration
+ * hierarchy, make sure CPU will wake up in time to handle
+ * remote timers. next_tmigr == KTIME_MAX if other CPUs are
+ * still active.
+ */
+ if (next_tmigr < tevt.local) {
+ u64 tmp;
+
+ /* If we missed a tick already, force 0 delta */
+ if (next_tmigr < basem)
+ next_tmigr = basem;
+
+ tmp = div_u64(next_tmigr - basem, TICK_NSEC);
+
+ nextevt = basej + (unsigned long)tmp;
+ tevt.local = next_tmigr;
+ *idle = time_after(nextevt, basej + 1);
+ }
+ /*
+ * Update of nextevt is not required in an else path, as it
+ * is revisited in !is_idle path only.
+ */
+ }
+
/* We need to mark both bases in sync */
base_local->is_idle = base_global->is_idle = *idle;
raw_spin_unlock(&base_global->lock);
raw_spin_unlock(&base_local->lock);
- return min_t(u64, tevt.local, tevt.global);
+ return tevt.local;
}
/**
@@ -2218,6 +2280,9 @@ void timer_clear_idle(void)
*/
__this_cpu_write(timer_bases[BASE_LOCAL].is_idle, false);
__this_cpu_write(timer_bases[BASE_GLOBAL].is_idle, false);
+
+ /* Activate without holding the timer_base->lock */
+ tmigr_cpu_activate();
}
#endif
@@ -2283,6 +2348,9 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h)
if (IS_ENABLED(CONFIG_NO_HZ_COMMON)) {
run_timer_base(BASE_GLOBAL);
run_timer_base(BASE_DEF);
+
+ if (is_timers_nohz_active())
+ tmigr_handle_remote();
}
}
@@ -2297,7 +2365,8 @@ static void run_local_timers(void)
for (int i = 0; i < NR_BASES; i++, base++) {
/* Raise the softirq only if required. */
- if (time_after_eq(jiffies, base->next_expiry)) {
+ if (time_after_eq(jiffies, base->next_expiry) ||
+ (i == BASE_DEF && tmigr_requires_handle_remote())) {
raise_softirq(TIMER_SOFTIRQ);
return;
}
diff --git a/kernel/time/timer_migration.c b/kernel/time/timer_migration.c
new file mode 100644
index 000000000000..929b3b94b893
--- /dev/null
+++ b/kernel/time/timer_migration.c
@@ -0,0 +1,1612 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Infrastructure for migratable timers
+ *
+ * Copyright(C) 2022 linutronix GmbH
+ */
+#include <linux/cpuhotplug.h>
+#include <linux/slab.h>
+#include <linux/smp.h>
+#include <linux/spinlock.h>
+#include <linux/timerqueue.h>
+#include <trace/events/ipi.h>
+
+#include "timer_migration.h"
+#include "tick-internal.h"
+
+/*
+ * The timer migration mechanism is built on a hierarchy of groups. The
+ * lowest level group contains CPUs, the next level groups of CPU groups
+ * and so forth. The CPU groups are kept per node so for the normal case
+ * lock contention won't happen across nodes. Depending on the number of
+ * CPUs per node even the next level might be kept as groups of CPU groups
+ * per node and only the levels above cross the node topology.
+ *
+ * Example topology for a two node system with 24 CPUs each.
+ *
+ * LVL 2 [GRP2:0]
+ * GRP1:0 = GRP1:M
+ *
+ * LVL 1 [GRP1:0] [GRP1:1]
+ * GRP0:0 - GRP0:2 GRP0:3 - GRP0:5
+ *
+ * LVL 0 [GRP0:0] [GRP0:1] [GRP0:2] [GRP0:3] [GRP0:4] [GRP0:5]
+ * CPUS 0-7 8-15 16-23 24-31 32-39 40-47
+ *
+ * The groups hold a timer queue of events sorted by expiry time. These
+ * queues are updated when CPUs go in idle. When they come out of idle
+ * ignore flag of events is set.
+ *
+ * Each group has a designated migrator CPU/group as long as a CPU/group is
+ * active in the group. This designated role is necessary to avoid that all
+ * active CPUs in a group try to migrate expired timers from other CPUs,
+ * which would result in massive lock bouncing.
+ *
+ * When a CPU is awake, it checks in it's own timer tick the group
+ * hierarchy up to the point where it is assigned the migrator role or if
+ * no CPU is active, it also checks the groups where no migrator is set
+ * (TMIGR_NONE).
+ *
+ * If it finds expired timers in one of the group queues it pulls them over
+ * from the idle CPU and runs the timer function. After that it updates the
+ * group and the parent groups if required.
+ *
+ * CPUs which go idle arm their CPU local timer hardware for the next local
+ * (pinned) timer event. If the next migratable timer expires after the
+ * next local timer or the CPU has no migratable timer pending then the
+ * CPU does not queue an event in the LVL0 group. If the next migratable
+ * timer expires before the next local timer then the CPU queues that timer
+ * in the LVL0 group. In both cases the CPU marks itself idle in the LVL0
+ * group.
+ *
+ * When CPU comes out of idle and when a group has at least a single active
+ * child, the ignore flag of the tmigr_event is set. This indicates, that
+ * the event is ignored even if it is still enqueued in the parent groups
+ * timer queue. It will be removed when touching the timer queue the next
+ * time. This spares locking in active path as the lock protects (after
+ * setup) only event information. For more information about locking,
+ * please read the section "Locking rules".
+ *
+ * If the CPU is the migrator of the group then it delegates that role to
+ * the next active CPU in the group or sets migrator to TMIGR_NONE when
+ * there is no active CPU in the group. This delegation needs to be
+ * propagated up the hierarchy so hand over from other leaves can happen at
+ * all hierarchy levels w/o doing a search.
+ *
+ * When the last CPU in the system goes idle, then it drops all migrator
+ * duties up to the top level of the hierarchy (LVL2 in the example). It
+ * then has to make sure, that it arms it's own local hardware timer for
+ * the earliest event in the system.
+ *
+ *
+ * Lifetime rules:
+ * ---------------
+ *
+ * The groups are built up at init time or when CPUs come online. They are
+ * not destroyed when a group becomes empty due to offlining. The group
+ * just won't participate in the hierarchy management anymore. Destroying
+ * groups would result in interesting race conditions which would just make
+ * the whole mechanism slow and complex.
+ *
+ *
+ * Locking rules:
+ * --------------
+ *
+ * For setting up new groups and handling events it's required to lock both
+ * child and parent group. The lock ordering is always bottom up. This also
+ * includes the per CPU locks in struct tmigr_cpu. For updating the migrator and
+ * active CPU/group information atomic_try_cmpxchg() is used instead and only
+ * the per CPU tmigr_cpu->lock is held.
+ *
+ * During the setup of groups tmigr_level_list is required. It is protected by
+ * @tmigr_mutex.
+ *
+ * When @timer_base->lock as well as tmigr related locks are required, the lock
+ * ordering is: first @timer_base->lock, afterwards tmigr related locks.
+ *
+ *
+ * Protection of the tmigr group state information:
+ * ------------------------------------------------
+ *
+ * The state information with the list of active children and migrator needs to
+ * be protected by a sequence counter. It prevents a race when updates in a
+ * child groups are propagated in changed order. The following scenario
+ * describes what happens without updating the sequence counter:
+ *
+ * Therefore, let's take three groups and four CPUs (CPU2 and CPU3 as well
+ * as GRP0:1 will not change during the scenario):
+ *
+ * LVL 1 [GRP1:0]
+ * migrator = GRP0:1
+ * active = GRP0:0, GRP0:1
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * migrator = CPU0 migrator = CPU2
+ * active = CPU0 active = CPU2
+ * / \ / \
+ * CPUs 0 1 2 3
+ * active idle active idle
+ *
+ *
+ * 1. CPU0 goes idle (changes are updated in GRP0:0; afterwards the current
+ * states of GRP0:0 and GRP1:0 are stored in the data for walking the
+ * hierarchy):
+ *
+ * LVL 1 [GRP1:0]
+ * migrator = GRP0:1
+ * active = GRP0:0, GRP0:1
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * --> migrator = TMIGR_NONE migrator = CPU2
+ * --> active = active = CPU2
+ * / \ / \
+ * CPUs 0 1 2 3
+ * --> idle idle active idle
+ *
+ * 2. CPU1 comes out of idle (changes are update in GRP0:0; afterwards the
+ * current states of GRP0:0 and GRP1:0 are stored in the data for walking the
+ * hierarchy):
+ *
+ * LVL 1 [GRP1:0]
+ * migrator = GRP0:1
+ * active = GRP0:0, GRP0:1
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * --> migrator = CPU1 migrator = CPU2
+ * --> active = CPU1 active = CPU2
+ * / \ / \
+ * CPUs 0 1 2 3
+ * idle --> active active idle
+ *
+ * 3. Here comes the change of the order: Propagating the changes of step 2
+ * through the hierarchy to GRP1:0 - nothing to be done, because GRP0:0
+ * is already up to date.
+ *
+ * 4. Propagating the changes of step 1 through the hierarchy to GRP1:0
+ *
+ * LVL 1 [GRP1:0]
+ * --> migrator = GRP0:1
+ * --> active = GRP0:1
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * migrator = CPU1 migrator = CPU2
+ * active = CPU1 active = CPU2
+ * / \ / \
+ * CPUs 0 1 2 3
+ * idle active active idle
+ *
+ * Now there is a inconsistent overall state because GRP0:0 is active, but
+ * it is marked as idle in the GRP1:0. This is prevented by incrementing
+ * sequence counter whenever changing the state.
+ *
+ *
+ * Required event and timerqueue update after a remote expiry:
+ * -----------------------------------------------------------
+ *
+ * After a remote expiry of a CPU, a walk through the hierarchy updating the
+ * events and timerqueues has to be done when there is a 'new' global timer of
+ * the remote CPU (which is obvious) but also if there is no new global timer,
+ * but the remote CPU is still idle:
+ *
+ * 1. CPU2 is the migrator and does the remote expiry in GRP1:0; expiry of
+ * evt-CPU0 and evt-CPU1 are equal:
+ *
+ * LVL 1 [GRP1:0]
+ * migrator = GRP0:1
+ * active = GRP0:1
+ * --> timerqueue = evt-GRP0:0
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * migrator = TMIGR_NONE migrator = CPU2
+ * active = active = CPU2
+ * groupevt.ignore = false groupevt.ignore = true
+ * groupevt.cpu = CPU0 groupevt.cpu =
+ * timerqueue = evt-CPU0, timerqueue =
+ * evt-CPU1
+ * / \ / \
+ * CPUs 0 1 2 3
+ * idle idle active idle
+ *
+ * 2. Remove the first event of the timerqueue in GRP1:0 and expire the timers
+ * of CPU0 (see evt-GRP0:0->cpu value):
+ *
+ * LVL 1 [GRP1:0]
+ * migrator = GRP0:1
+ * active = GRP0:1
+ * --> timerqueue =
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * migrator = TMIGR_NONE migrator = CPU2
+ * active = active = CPU2
+ * groupevt.ignore = false groupevt.ignore = true
+ * --> groupevt.cpu = CPU0 groupevt.cpu =
+ * timerqueue = evt-CPU0, timerqueue =
+ * evt-CPU1
+ * / \ / \
+ * CPUs 0 1 2 3
+ * idle idle active idle
+ *
+ * 3. After the remote expiry CPU0 has no global timer that needs to be
+ * enqueued. When skipping the walk, the global timer of CPU1 is not handled,
+ * as the group event of GRP0:0 is not updated and not enqueued into GRP1:0. The
+ * walk has to be done to update the group events and timerqueues:
+ *
+ * LVL 1 [GRP1:0]
+ * migrator = GRP0:1
+ * active = GRP0:1
+ * --> timerqueue = evt-GRP0:0
+ * / \
+ * LVL 0 [GRP0:0] [GRP0:1]
+ * migrator = TMIGR_NONE migrator = CPU2
+ * active = active = CPU2
+ * groupevt.ignore = false groupevt.ignore = true
+ * --> groupevt.cpu = CPU1 groupevt.cpu =
+ * --> timerqueue = evt-CPU1 timerqueue =
+ * / \ / \
+ * CPUs 0 1 2 3
+ * idle idle active idle
+ *
+ * Now CPU2 (migrator) is able to handle the timer of CPU1 as CPU2 only scans
+ * the timerqueues of GRP0:1 and GRP1:0.
+ *
+ * The update of step 3 is valid to be skipped, when the remote CPU went offline
+ * in the meantime because an update was already done during inactive path. When
+ * CPU became active in the meantime, update isn't required as well, because
+ * GRP0:0 is now longer idle.
+ */
+
+static DEFINE_MUTEX(tmigr_mutex);
+static struct list_head *tmigr_level_list __read_mostly;
+
+static unsigned int tmigr_hierarchy_levels __read_mostly;
+static unsigned int tmigr_crossnode_level __read_mostly;
+
+static DEFINE_PER_CPU(struct tmigr_cpu, tmigr_cpu);
+
+#define TMIGR_NONE 0xFF
+#define BIT_CNT 8
+
+static DEFINE_STATIC_KEY_FALSE(tmigr_enabled);
+
+static inline bool is_tmigr_enabled(void)
+{
+ return static_branch_unlikely(&tmigr_enabled);
+}
+
+/*
+ * Returns true, when @childmask corresponds to the group migrator or when the
+ * group is not active - so no migrator is set.
+ */
+static bool tmigr_check_migrator(struct tmigr_group *group, u8 childmask)
+{
+ union tmigr_state s;
+
+ s.state = atomic_read(&group->migr_state);
+
+ if ((s.migrator == childmask) || (s.migrator == TMIGR_NONE))
+ return true;
+
+ return false;
+}
+
+static bool tmigr_check_migrator_and_lonely(struct tmigr_group *group, u8 childmask)
+{
+ bool lonely, migrator = false;
+ unsigned long active;
+ union tmigr_state s;
+
+ s.state = atomic_read(&group->migr_state);
+
+ if ((s.migrator == childmask) || (s.migrator == TMIGR_NONE))
+ migrator = true;
+
+ active = s.active;
+ lonely = bitmap_weight(&active, BIT_CNT) <= 1;
+
+ return (migrator && lonely);
+}
+
+static bool tmigr_check_lonely(struct tmigr_group *group)
+{
+ unsigned long active;
+ union tmigr_state s;
+
+ s.state = atomic_read(&group->migr_state);
+
+ active = s.active;
+
+ return bitmap_weight(&active, BIT_CNT) <= 1;
+}
+
+typedef bool (*up_f)(struct tmigr_group *, struct tmigr_group *, void *);
+
+static void __walk_groups(up_f up, void *data,
+ struct tmigr_cpu *tmc)
+{
+ struct tmigr_group *child = NULL, *group = tmc->tmgroup;
+
+ do {
+ WARN_ON_ONCE(group->level >= tmigr_hierarchy_levels);
+
+ if (up(group, child, data))
+ break;
+
+ child = group;
+ group = group->parent;
+ } while (group);
+}
+
+static void walk_groups(up_f up, void *data, struct tmigr_cpu *tmc)
+{
+ lockdep_assert_held(&tmc->lock);
+
+ __walk_groups(up, data, tmc);
+}
+
+/**
+ * struct tmigr_walk - data required for walking the hierarchy
+ * @evt: Pointer to tmigr_event which needs to be queued (of idle
+ * child group)
+ * @childmask: childmask of child group
+ * @nextexp: Next CPU event expiry information which is handed into
+ * the timer migration code by the timer code
+ * (get_next_timer_interrupt()); it is furthermore used for
+ * the first event which is queued, if the timer migration
+ * hierarchy is completely idle
+ * @childstate: tmigr_group->migr_state of the child - will be only
+ * reread when cmpxchg in the group fails (is required for
+ * the deactivate path and the new timer path)
+ * @groupstate: tmigr_group->migr_state of the group - will be only
+ * reread when cmpxchg in the group fails (is required for
+ * the active, the deactivate and the new timer path)
+ * @remote: Is set, when the new timer path is executed in
+ * tmigr_handle_remote_cpu()
+ */
+struct tmigr_walk {
+ struct tmigr_event *evt;
+ u8 childmask;
+ u64 nextexp;
+ union tmigr_state childstate;
+ union tmigr_state groupstate;
+ bool remote;
+};
+
+/**
+ * struct tmigr_remote_data - data required for (check) remote expiry
+ * hierarchy walk
+ * @basej: timer base in jiffies
+ * @now: timer base monotonic
+ * @nextexp: returns expiry of the first timer in the idle timer
+ * migration hierarchy to make sure the timer is handled in
+ * time; it is stored in the per CPU tmigr_cpu struct of
+ * CPU which expires remote timers
+ * @childmask: childmask of child group
+ * @check: is set if there is the need to handle remote timers;
+ * required in tmigr_check_handle_remote() only
+ * @tmc_active: this flag indicates, whether the CPU which triggers
+ * the hierarchy walk is !idle in the timer migration
+ * hierarchy. When the CPU is idle and the whole hierarchy is
+ * idle, only the first event of the top level has to be
+ * considered.
+ */
+struct tmigr_remote_data {
+ unsigned long basej;
+ u64 now;
+ u64 nextexp;
+ u8 childmask;
+ bool check;
+ bool tmc_active;
+};
+
+/*
+ * Returns the next event of the timerqueue @group->events
+ *
+ * Removes timers with ignore flag and update next_expiry of the group. Values
+ * of the group event are updated in tmigr_update_events() only.
+ */
+static struct tmigr_event *tmigr_next_groupevt(struct tmigr_group *group)
+{
+ struct timerqueue_node *node = NULL;
+ struct tmigr_event *evt = NULL;
+
+ lockdep_assert_held(&group->lock);
+
+ WRITE_ONCE(group->next_expiry, KTIME_MAX);
+
+ while ((node = timerqueue_getnext(&group->events))) {
+ evt = container_of(node, struct tmigr_event, nextevt);
+
+ if (!evt->ignore) {
+ WRITE_ONCE(group->next_expiry, evt->nextevt.expires);
+ return evt;
+ }
+
+ /*
+ * Remove next timers with ignore flag, because the group lock
+ * is held anyway
+ */
+ if (!timerqueue_del(&group->events, node))
+ break;
+ }
+
+ return NULL;
+}
+
+/*
+ * Return the next event which is already expired of the group timerqueue
+ *
+ * Event, which is returned, is also removed from the queue.
+ */
+static struct tmigr_event *tmigr_next_expired_groupevt(struct tmigr_group *group,
+ u64 now)
+{
+ struct tmigr_event *evt = tmigr_next_groupevt(group);
+
+ if (!evt || now < evt->nextevt.expires)
+ return NULL;
+
+ /*
+ * The event is already expired. Remove it. If it's not the last event,
+ * then update all group event related information.
+ */
+ if (timerqueue_del(&group->events, &evt->nextevt))
+ tmigr_next_groupevt(group);
+ else
+ WRITE_ONCE(group->next_expiry, KTIME_MAX);
+
+ return evt;
+}
+
+static u64 tmigr_next_groupevt_expires(struct tmigr_group *group)
+{
+ struct tmigr_event *evt;
+
+ evt = tmigr_next_groupevt(group);
+
+ if (!evt)
+ return KTIME_MAX;
+ else
+ return evt->nextevt.expires;
+}
+
+static bool tmigr_active_up(struct tmigr_group *group,
+ struct tmigr_group *child,
+ void *ptr)
+{
+ union tmigr_state curstate, newstate;
+ struct tmigr_walk *data = ptr;
+ bool walk_done;
+ u8 childmask;
+
+ childmask = data->childmask;
+ newstate = curstate = data->groupstate;
+
+retry:
+ walk_done = true;
+
+ if (newstate.migrator == TMIGR_NONE) {
+ newstate.migrator = childmask;
+
+ /* Changes need to be propagated */
+ walk_done = false;
+ }
+
+ newstate.active |= childmask;
+
+ newstate.seq++;
+
+ if (!atomic_try_cmpxchg(&group->migr_state, &curstate.state, newstate.state)) {
+ newstate.state = curstate.state;
+ goto retry;
+ }
+
+ if (group->parent && (walk_done == false)) {
+ data->groupstate.state = atomic_read(&group->parent->migr_state);
+ data->childmask = group->childmask;
+ }
+
+ /*
+ * The group is active and the event will be ignored - the ignore flag is
+ * updated without holding the lock. In case the bit is set while
+ * another CPU already handles remote events, nothing happens, because
+ * it is clear that the CPU became active just in this moment, or in
+ * worst case the event is handled remote. Nothing to worry about.
+ */
+ group->groupevt.ignore = true;
+
+ return walk_done;
+}
+
+static void __tmigr_cpu_activate(struct tmigr_cpu *tmc)
+{
+ struct tmigr_walk data;
+
+ data.childmask = tmc->childmask;
+ data.groupstate.state = atomic_read(&tmc->tmgroup->migr_state);
+
+ tmc->cpuevt.ignore = true;
+ WRITE_ONCE(tmc->wakeup, KTIME_MAX);
+ tmc->wakeup_recalc = false;
+
+ walk_groups(&tmigr_active_up, &data, tmc);
+}
+
+/**
+ * tmigr_cpu_activate - set CPU active in timer migration hierarchy
+ *
+ * Call site timer_clear_idle() is called with interrupts disabled
+ */
+void tmigr_cpu_activate(void)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+
+ if (!is_tmigr_enabled() || !tmc->tmgroup || !tmc->online || !tmc->idle)
+ return;
+
+ raw_spin_lock(&tmc->lock);
+ tmc->idle = false;
+ __tmigr_cpu_activate(tmc);
+ raw_spin_unlock(&tmc->lock);
+}
+
+/*
+ * Returns true, if there is nothing to be propagated to the next level
+ *
+ * @data->nextexp is reset to KTIME_MAX; it is reused for the first global event
+ * which needs to be handled by the migrator (in the top level group).
+ *
+ * This is the only place where the group event expiry value is set.
+ */
+static bool tmigr_update_events(struct tmigr_group *group,
+ struct tmigr_group *child,
+ struct tmigr_walk *data)
+{
+ struct tmigr_event *evt, *first_childevt;
+ bool walk_done, remote = data->remote;
+ bool leftmost_change = false;
+ u64 nextexp;
+
+ if (child) {
+ raw_spin_lock(&child->lock);
+ raw_spin_lock_nested(&group->lock, SINGLE_DEPTH_NESTING);
+
+ data->childstate.state = atomic_read(&child->migr_state);
+ data->groupstate.state = atomic_read(&group->migr_state);
+
+ if (data->childstate.active) {
+ walk_done = true;
+ goto unlock;
+ }
+
+ first_childevt = tmigr_next_groupevt(child);
+ nextexp = child->next_expiry;
+ evt = &child->groupevt;
+ } else {
+ nextexp = data->nextexp;
+
+ /*
+ * Set @data->nextexp to KTIME_MAX; it is reused for the first
+ * global event which needs to be handled by the migrator (in
+ * the top level group).
+ */
+ data->nextexp = KTIME_MAX;
+
+ first_childevt = evt = data->evt;
+
+ /*
+ * Walking the hierarchy is required in any case when a
+ * remote expiry was done before. This ensures to not lose
+ * already queued events in non active groups (see section
+ * "Required event and timerqueue update after remote
+ * expiry" in the documentation at the top).
+ *
+ * The two call sites which are executed without a remote expiry
+ * before, are not prevented from propagating changes through
+ * the hierarchy by the return:
+ * - When entering this path by tmigr_new_timer(), @evt->ignore
+ * is never set.
+ * - tmigr_inactive_up() takes care of the propagation by
+ * itself and ignores the return value. But an immediate
+ * return is required because nothing has to be done in this
+ * level as the event could be ignored.
+ */
+ if (evt->ignore && !remote)
+ return true;
+
+ raw_spin_lock(&group->lock);
+ data->groupstate.state = atomic_read(&group->migr_state);
+ }
+
+ if (nextexp == KTIME_MAX) {
+ evt->ignore = true;
+
+ /*
+ * When the next child event could be ignored (nextexp is
+ * KTIME_MAX) and there was no remote timer handling before or
+ * the group is already active, there is no need to walk the
+ * hierarchy even if there is a parent group.
+ *
+ * The other way round: even if the event could be ignored, but
+ * if a remote timer handling was executed before and the group
+ * is not active, walking the hierarchy is required to not miss
+ * an enqueued timer in the non active group. The enqueued timer
+ * of the group needs to be propagated to a higher level to
+ * ensure it is handled.
+ */
+ if (!remote || data->groupstate.active) {
+ walk_done = true;
+ goto unlock;
+ }
+ } else {
+ /*
+ * An update of @evt->cpu and @evt->ignore flag is required only
+ * when @child is set (the child is equal or higher than lvl0),
+ * but it doesn't matter if it is written once more to the per
+ * CPU event; make the update unconditional.
+ */
+ evt->cpu = first_childevt->cpu;
+ evt->ignore = false;
+ }
+
+ walk_done = !group->parent;
+
+ /*
+ * If the child event is already queued in the group, remove it from the
+ * queue when the expiry time changed only.
+ */
+ if (timerqueue_node_queued(&evt->nextevt)) {
+ if (evt->nextevt.expires == nextexp)
+ goto check_toplvl;
+
+ leftmost_change = timerqueue_getnext(&group->events) == &evt->nextevt;
+ if (!timerqueue_del(&group->events, &evt->nextevt))
+ WRITE_ONCE(group->next_expiry, KTIME_MAX);
+ }
+
+ evt->nextevt.expires = nextexp;
+
+ if (timerqueue_add(&group->events, &evt->nextevt)) {
+ leftmost_change = true;
+ WRITE_ONCE(group->next_expiry, nextexp);
+ }
+
+check_toplvl:
+ if (walk_done && (data->groupstate.migrator == TMIGR_NONE)) {
+ /*
+ * Nothing to do when first event didn't changed and update was
+ * done during remote timer handling.
+ */
+ if (remote && !leftmost_change)
+ goto unlock;
+ /*
+ * The top level group is idle and it has to be ensured the
+ * global timers are handled in time. (This could be optimized
+ * by keeping track of the last global scheduled event and only
+ * arming it on the CPU if the new event is earlier. Not sure if
+ * its worth the complexity.)
+ */
+ data->nextexp = tmigr_next_groupevt_expires(group);
+ }
+
+unlock:
+ raw_spin_unlock(&group->lock);
+
+ if (child)
+ raw_spin_unlock(&child->lock);
+
+ return walk_done;
+}
+
+static bool tmigr_new_timer_up(struct tmigr_group *group,
+ struct tmigr_group *child,
+ void *ptr)
+{
+ struct tmigr_walk *data = ptr;
+
+ return tmigr_update_events(group, child, data);
+}
+
+/*
+ * Returns the expiry of the next timer that needs to be handled. KTIME_MAX is
+ * returned, when an active CPU will handle all the timer migration hierarchy
+ * timers.
+ */
+static u64 tmigr_new_timer(struct tmigr_cpu *tmc, u64 nextexp)
+{
+ struct tmigr_walk data = { .evt = &tmc->cpuevt,
+ .nextexp = nextexp };
+
+ lockdep_assert_held(&tmc->lock);
+
+ if (tmc->remote)
+ return KTIME_MAX;
+
+ tmc->cpuevt.ignore = false;
+ data.remote = false;
+
+ walk_groups(&tmigr_new_timer_up, &data, tmc);
+
+ /* If there is a new first global event, make sure it is handled */
+ return data.nextexp;
+}
+
+static u64 tmigr_handle_remote_cpu(unsigned int cpu, u64 now,
+ unsigned long jif)
+{
+ struct timer_events tevt;
+ struct tmigr_walk data;
+ struct tmigr_cpu *tmc;
+ u64 next = KTIME_MAX;
+
+ tmc = per_cpu_ptr(&tmigr_cpu, cpu);
+
+ raw_spin_lock_irq(&tmc->lock);
+
+ /*
+ * The remote CPU is offline or the CPU event does not has to be handled
+ * (the CPU is active or there is no longer an event to expire) or
+ * another CPU handles the CPU timers already or the next event was
+ * already expired - return!
+ */
+ if (!tmc->online || tmc->remote || tmc->cpuevt.ignore ||
+ now < tmc->cpuevt.nextevt.expires) {
+ raw_spin_unlock_irq(&tmc->lock);
+ return next;
+ }
+
+ tmc->remote = true;
+ WRITE_ONCE(tmc->wakeup, KTIME_MAX);
+
+ /* Drop the lock to allow the remote CPU to exit idle */
+ raw_spin_unlock_irq(&tmc->lock);
+
+ if (cpu != smp_processor_id())
+ timer_expire_remote(cpu);
+
+ /*
+ * Lock ordering needs to be preserved - timer_base locks before tmigr
+ * related locks (see section "Locking rules" in the documentation at
+ * the top). During fetching the next timer interrupt, also tmc->lock
+ * needs to be held. Otherwise there is a possible race window against
+ * the CPU itself when it comes out of idle, updates the first timer in
+ * the hierarchy and goes back to idle.
+ *
+ * timer base locks are dropped as fast as possible: After checking
+ * whether the remote CPU went offline in the meantime and after
+ * fetching the next remote timer interrupt. Dropping the locks as fast
+ * as possible keeps the locking region small and prevents holding
+ * several (unnecessary) locks during walking the hierarchy for updating
+ * the timerqueue and group events.
+ */
+ local_irq_disable();
+ timer_lock_remote_bases(cpu);
+ raw_spin_lock(&tmc->lock);
+
+ /*
+ * When the CPU went offline in the meantime, no hierarchy walk has to
+ * be done for updating the queued events, because the walk was
+ * already done during marking the CPU offline in the hierarchy.
+ *
+ * When the CPU is no longer idle, the CPU takes care of the timers and
+ * also of the timers in the path to the top.
+ *
+ * (See also section "Required event and timerqueue update after
+ * remote expiry" in the documentation at the top)
+ */
+ if (!tmc->online || !tmc->idle) {
+ timer_unlock_remote_bases(cpu);
+ goto unlock;
+ } else {
+ /* next event of CPU */
+ fetch_next_timer_interrupt_remote(jif, now, &tevt, cpu);
+ }
+
+ timer_unlock_remote_bases(cpu);
+
+ data.evt = &tmc->cpuevt;
+ data.nextexp = tevt.global;
+ data.remote = true;
+
+ /*
+ * The update is done even when there is no 'new' global timer pending
+ * on the remote CPU (see section "Required event and timerqueue update
+ * after remote expiry" in the documentation at the top)
+ */
+ walk_groups(&tmigr_new_timer_up, &data, tmc);
+
+ next = data.nextexp;
+
+unlock:
+ tmc->remote = false;
+ raw_spin_unlock_irq(&tmc->lock);
+
+ return next;
+}
+
+static bool tmigr_handle_remote_up(struct tmigr_group *group,
+ struct tmigr_group *child,
+ void *ptr)
+{
+ struct tmigr_remote_data *data = ptr;
+ u64 now, next = KTIME_MAX;
+ struct tmigr_event *evt;
+ unsigned long jif;
+ u8 childmask;
+
+ jif = data->basej;
+ now = data->now;
+
+ childmask = data->childmask;
+
+again:
+ /*
+ * Handle the group only if @childmask is the migrator or if the
+ * group has no migrator. Otherwise the group is active and is
+ * handled by its own migrator.
+ */
+ if (!tmigr_check_migrator(group, childmask))
+ return true;
+
+ raw_spin_lock_irq(&group->lock);
+
+ evt = tmigr_next_expired_groupevt(group, now);
+
+ if (evt) {
+ unsigned int remote_cpu = evt->cpu;
+
+ raw_spin_unlock_irq(&group->lock);
+
+ next = tmigr_handle_remote_cpu(remote_cpu, now, jif);
+
+ /* check if there is another event, that needs to be handled */
+ goto again;
+ } else {
+ raw_spin_unlock_irq(&group->lock);
+ }
+
+ /* Update of childmask for the next level */
+ data->childmask = group->childmask;
+ data->nextexp = next;
+
+ return false;
+}
+
+/**
+ * tmigr_handle_remote - Handle migratable timers on remote idle CPUs
+ *
+ * Called from the timer soft interrupt with interrupts enabled.
+ */
+void tmigr_handle_remote(void)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+ struct tmigr_remote_data data;
+
+ if (!is_tmigr_enabled() || !tmc->tmgroup || !tmc->online)
+ return;
+
+ data.childmask = tmc->childmask;
+ data.nextexp = KTIME_MAX;
+
+ /*
+ * NOTE: This is a doubled check because the migrator test will be done
+ * in tmigr_handle_remote_up() anyway. Keep this check to fasten the
+ * return when nothing has to be done.
+ */
+ if (!tmigr_check_migrator(tmc->tmgroup, tmc->childmask))
+ return;
+
+ data.now = get_jiffies_update(&data.basej);
+
+ /*
+ * Update @tmc->wakeup only at the end and do not reset @tmc->wakeup to
+ * KTIME_MAX. Even if tmc->lock is not held during the whole remote
+ * handling, tmc->wakeup is fine to be stale as it is called in
+ * interrupt context and tick_nohz_next_event() is executed in interrupt
+ * exit path only after processing the last pending interrupt.
+ */
+
+ __walk_groups(&tmigr_handle_remote_up, &data, tmc);
+
+ raw_spin_lock_irq(&tmc->lock);
+ WRITE_ONCE(tmc->wakeup, data.nextexp);
+ raw_spin_unlock_irq(&tmc->lock);
+}
+
+static bool tmigr_requires_handle_remote_up(struct tmigr_group *group,
+ struct tmigr_group *child,
+ void *ptr)
+{
+ struct tmigr_remote_data *data = ptr;
+ u8 childmask;
+
+ childmask = data->childmask;
+
+ /*
+ * Handle the group only if the child is the migrator or if the group
+ * has no migrator. Otherwise the group is active and is handled by its
+ * own migrator.
+ */
+ if (!tmigr_check_migrator(group, childmask))
+ return true;
+
+ /*
+ * When there is a parent group and the CPU which triggered the
+ * hierarchy walk is not active, proceed the walk to reach the top level
+ * group before reading the next_expiry value.
+ */
+ if (group->parent && !data->tmc_active)
+ goto out;
+
+ /*
+ * On 32 bit systems the racy lockless check for next_expiry will
+ * turn into a random number generator. Therefore do the lockless
+ * check only on 64 bit systems.
+ */
+ if (IS_ENABLED(CONFIG_64BIT)) {
+ data->nextexp = READ_ONCE(group->next_expiry);
+ if (data->now >= data->nextexp) {
+ data->check = true;
+ return true;
+ }
+ } else {
+ raw_spin_lock(&group->lock);
+ data->nextexp = group->next_expiry;
+ if (data->now >= group->next_expiry) {
+ data->check = true;
+ raw_spin_unlock(&group->lock);
+ return true;
+ }
+ raw_spin_unlock(&group->lock);
+ }
+
+out:
+ /* Update of childmask for the next level */
+ data->childmask = group->childmask;
+ return false;
+}
+
+/**
+ * tmigr_requires_handle_remote - Check whether remote timer handling is required
+ *
+ * Must be called with interrupts disabled.
+ */
+int tmigr_requires_handle_remote(void)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+ struct tmigr_remote_data data;
+ unsigned int ret = 0;
+ unsigned long jif;
+
+ if (!is_tmigr_enabled() || !tmc->tmgroup || !tmc->online)
+ return ret;
+
+ data.now = get_jiffies_update(&jif);
+ data.childmask = tmc->childmask;
+ data.nextexp = KTIME_MAX;
+ data.tmc_active = !tmc->idle;
+ data.check = false;
+
+ /*
+ * When the CPU is active, walking the hierarchy to check whether a
+ * remote expiry is required.
+ *
+ * Check is done lockless as interrupts are disabled and @tmc->idle is
+ * set only by the local CPU.
+ */
+ if (!tmc->idle) {
+ __walk_groups(&tmigr_requires_handle_remote_up, &data, tmc);
+
+ if (data.nextexp != KTIME_MAX)
+ ret = 1;
+
+ return ret;
+ }
+
+ /*
+ * When the CPU is idle, check whether the recalculation of @tmc->wakeup
+ * is required. @tmc->wakeup_recalc is set by a remote CPU which is
+ * about to go offline, was the last active CPU in the whole timer
+ * migration hierarchy and now delegates handling of the hierarchy to
+ * this CPU.
+ *
+ * Racy lockless check is valid:
+ * - @tmc->wakeup_recalc is set by the remote CPU before it issues
+ * reschedule IPI.
+ * - As interrupts are disabled here this CPU will either observe
+ * @tmc->wakeup_recalc set before the reschedule IPI can be handled or
+ * it will observe it when this function is called again on return
+ * from handling the reschedule IPI.
+ */
+ if (tmc->wakeup_recalc) {
+ raw_spin_lock(&tmc->lock);
+
+ __walk_groups(&tmigr_requires_handle_remote_up, &data, tmc);
+
+ if (data.nextexp != KTIME_MAX)
+ ret = 1;
+
+ WRITE_ONCE(tmc->wakeup, data.nextexp);
+ tmc->wakeup_recalc = false;
+ raw_spin_unlock(&tmc->lock);
+
+ return ret;
+ }
+
+ /*
+ * When the CPU is idle and @tmc->wakeup is reliable, compare it with
+ * @data.now. On 64 bit it is valid to do this lockless. On 32 bit
+ * systems, holding the lock is required to get valid data on concurrent
+ * writers.
+ */
+ if (IS_ENABLED(CONFIG_64BIT)) {
+ if (data.now >= READ_ONCE(tmc->wakeup))
+ ret = 1;
+ } else {
+ raw_spin_lock(&tmc->lock);
+ if (data.now >= tmc->wakeup)
+ ret = 1;
+ raw_spin_unlock(&tmc->lock);
+ }
+
+ return ret;
+}
+
+static bool tmigr_inactive_up(struct tmigr_group *group,
+ struct tmigr_group *child,
+ void *ptr)
+{
+ union tmigr_state curstate, newstate;
+ struct tmigr_walk *data = ptr;
+ bool walk_done;
+ u8 childmask;
+
+ childmask = data->childmask;
+ newstate = curstate = data->groupstate;
+
+retry:
+ walk_done = true;
+
+ /* Reset active bit when the child is no longer active */
+ if (!data->childstate.active)
+ newstate.active &= ~childmask;
+
+ if (newstate.migrator == childmask) {
+ /*
+ * Find a new migrator for the group, because the child group is
+ * idle!
+ */
+ if (!data->childstate.active) {
+ unsigned long new_migr_bit, active = newstate.active;
+
+ new_migr_bit = find_first_bit(&active, BIT_CNT);
+
+ if (new_migr_bit != BIT_CNT) {
+ newstate.migrator = BIT(new_migr_bit);
+ } else {
+ newstate.migrator = TMIGR_NONE;
+
+ /* Changes need to be propagated */
+ walk_done = false;
+ }
+ }
+ }
+
+ newstate.seq++;
+
+ WARN_ON_ONCE((newstate.migrator != TMIGR_NONE) && !(newstate.active));
+
+ if (!atomic_try_cmpxchg(&group->migr_state, &curstate.state, newstate.state)) {
+ newstate.state = curstate.state;
+
+ /*
+ * Something changed in the child/parent group in the meantime,
+ * reread the state of the child and parent; Update of
+ * data->childstate is required for event handling;
+ */
+ if (child)
+ data->childstate.state = atomic_read(&child->migr_state);
+
+ goto retry;
+ }
+
+ data->groupstate = newstate;
+ data->remote = false;
+
+ /* Event Handling */
+ tmigr_update_events(group, child, data);
+
+ if (group->parent && (walk_done == false)) {
+ data->childmask = group->childmask;
+ data->childstate = newstate;
+ data->groupstate.state = atomic_read(&group->parent->migr_state);
+ }
+
+ /*
+ * data->nextexp was set by tmigr_update_events() and contains the
+ * expiry of the first global event which needs to be handled
+ */
+ if (data->nextexp != KTIME_MAX) {
+ WARN_ON_ONCE(group->parent);
+ /*
+ * Top level path - If this CPU is about going offline, wake
+ * up some random other CPU so it will take over the
+ * migrator duty and program its timer properly. Ideally
+ * wake the CPU with the closest expiry time, but that's
+ * overkill to figure out.
+ *
+ * Set wakeup_recalc of remote CPU, to make sure the complete
+ * idle hierarchy with enqueued timers is reevaluated.
+ */
+ if (!(this_cpu_ptr(&tmigr_cpu)->online)) {
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+ unsigned int cpu = smp_processor_id();
+ struct tmigr_cpu *tmc_resched;
+
+ cpu = cpumask_any_but(cpu_online_mask, cpu);
+ tmc_resched = per_cpu_ptr(&tmigr_cpu, cpu);
+
+ raw_spin_unlock(&tmc->lock);
+
+ raw_spin_lock(&tmc_resched->lock);
+ tmc_resched->wakeup_recalc = true;
+ raw_spin_unlock(&tmc_resched->lock);
+
+ raw_spin_lock(&tmc->lock);
+ smp_send_reschedule(cpu);
+ }
+ }
+
+ return walk_done;
+}
+
+static u64 __tmigr_cpu_deactivate(struct tmigr_cpu *tmc, u64 nextexp)
+{
+ struct tmigr_walk data = { .childmask = tmc->childmask,
+ .evt = &tmc->cpuevt,
+ .nextexp = nextexp,
+ .childstate.state = 0 };
+
+ data.groupstate.state = atomic_read(&tmc->tmgroup->migr_state);
+
+ /*
+ * If nextexp is KTIME_MAX, the CPU event will be ignored because the
+ * local timer expires before the global timer, no global timer is set
+ * or CPU goes offline.
+ */
+ if (nextexp != KTIME_MAX)
+ tmc->cpuevt.ignore = false;
+
+ walk_groups(&tmigr_inactive_up, &data, tmc);
+ return data.nextexp;
+}
+
+/**
+ * tmigr_cpu_deactivate - Put current CPU into inactive state
+ * @nextexp: The next timer event expiry set in the current CPU
+ *
+ * Must be called with interrupts disabled.
+ *
+ * Return: the next event expiry of the current CPU or the next event expiry
+ * from the hierarchy if this CPU is the top level migrator or the hierarchy is
+ * completely idle.
+ */
+u64 tmigr_cpu_deactivate(u64 nextexp)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+ u64 ret;
+
+ if (!is_tmigr_enabled() || !tmc->tmgroup || !tmc->online)
+ return nextexp;
+
+ raw_spin_lock(&tmc->lock);
+
+ /*
+ * The CPU is already deactivated in the timer migration
+ * hierarchy. tick_nohz_get_sleep_length() calls tick_nohz_next_event()
+ * and thereby the timer idle path is executed once more. @tmc->wakeup
+ * holds the first timer, when the timer migration hierarchy is
+ * completely idle. If there is no new next expiry value handed in which
+ * should be inserted into the timer migration hierarchy, the wakeup
+ * time is returned.
+ */
+ if (tmc->idle) {
+ ret = READ_ONCE(tmc->wakeup);
+ if (nextexp != KTIME_MAX) {
+ if (nextexp != tmc->cpuevt.nextevt.expires ||
+ tmc->cpuevt.ignore) {
+ ret = tmigr_new_timer(tmc, nextexp);
+ }
+ } else if (tmc->wakeup_recalc) {
+ struct tmigr_remote_data data;
+
+ data.now = KTIME_MAX;
+ data.childmask = tmc->childmask;
+ data.nextexp = KTIME_MAX;
+ data.tmc_active = false;
+ data.check = false;
+
+ __walk_groups(&tmigr_requires_handle_remote_up, &data, tmc);
+
+ ret = data.nextexp;
+ }
+ tmc->wakeup_recalc = false;
+ goto unlock;
+ }
+
+ ret = __tmigr_cpu_deactivate(tmc, nextexp);
+
+ tmc->idle = true;
+
+unlock:
+ /*
+ * Make sure the reevaluation of timers in idle path will not miss an
+ * event.
+ */
+ WRITE_ONCE(tmc->wakeup, ret);
+
+ raw_spin_unlock(&tmc->lock);
+ return ret;
+}
+
+/**
+ * tmigr_quick_check - Quick check which is called by tick_nohz_next_event
+ *
+ * Returns KTIME_MAX, when it is probable that nothing has to be done (not the
+ * only one in the level 0 group; and if it is the only one in level 0 group,
+ * but there are more than a single group active in top level)
+ *
+ * Returns first expiry of the top level group, when it is the only one in level
+ * 0 and top level also only has a single active child.
+ */
+u64 tmigr_quick_check(void)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+ struct tmigr_group *topgroup;
+ struct list_head lvllist;
+
+ if (!is_tmigr_enabled() || !tmc->tmgroup || !tmc->online || tmc->idle)
+ return KTIME_MAX;
+
+ if (!tmigr_check_migrator_and_lonely(tmc->tmgroup, tmc->childmask))
+ return KTIME_MAX;
+
+ for (int i = tmigr_hierarchy_levels; i > 0 ; i--) {
+ lvllist = tmigr_level_list[i - 1];
+ if (list_is_singular(&lvllist)) {
+ topgroup = list_first_entry(&lvllist, struct tmigr_group, list);
+
+ if (tmigr_check_lonely(topgroup))
+ return READ_ONCE(topgroup->next_expiry);
+ } else {
+ continue;
+ }
+ }
+
+ return KTIME_MAX;
+}
+
+static void tmigr_init_group(struct tmigr_group *group, unsigned int lvl,
+ int node)
+{
+ union tmigr_state s;
+
+ raw_spin_lock_init(&group->lock);
+
+ group->level = lvl;
+ group->numa_node = lvl < tmigr_crossnode_level ? node : NUMA_NO_NODE;
+
+ group->num_children = 0;
+
+ s.migrator = TMIGR_NONE;
+ s.active = 0;
+ s.seq = 0;
+ atomic_set(&group->migr_state, s.state);
+
+ timerqueue_init_head(&group->events);
+ timerqueue_init(&group->groupevt.nextevt);
+ group->groupevt.nextevt.expires = KTIME_MAX;
+ WRITE_ONCE(group->next_expiry, KTIME_MAX);
+ group->groupevt.ignore = true;
+}
+
+static struct tmigr_group *tmigr_get_group(unsigned int cpu, int node,
+ unsigned int lvl)
+{
+ struct tmigr_group *tmp, *group = NULL;
+
+ lockdep_assert_held(&tmigr_mutex);
+
+ /* Try to attach to an existing group first */
+ list_for_each_entry(tmp, &tmigr_level_list[lvl], list) {
+ /*
+ * If @lvl is below the cross numa node level, check whether
+ * this group belongs to the same numa node.
+ */
+ if (lvl < tmigr_crossnode_level && tmp->numa_node != node)
+ continue;
+
+ /* Capacity left? */
+ if (tmp->num_children >= TMIGR_CHILDREN_PER_GROUP)
+ continue;
+
+ /*
+ * TODO: A possible further improvement: Make sure that all CPU
+ * siblings end up in the same group of the lowest level of the
+ * hierarchy. Rely on the topology sibling mask would be a
+ * reasonable solution.
+ */
+
+ group = tmp;
+ break;
+ }
+
+ if (group)
+ return group;
+
+ /* Allocate and set up a new group */
+ group = kzalloc_node(sizeof(*group), GFP_KERNEL, node);
+ if (!group)
+ return ERR_PTR(-ENOMEM);
+
+ tmigr_init_group(group, lvl, node);
+
+ /* Setup successful. Add it to the hierarchy */
+ list_add(&group->list, &tmigr_level_list[lvl]);
+ return group;
+}
+
+static void tmigr_connect_child_parent(struct tmigr_group *child,
+ struct tmigr_group *parent)
+{
+ union tmigr_state childstate;
+
+ raw_spin_lock_irq(&child->lock);
+ raw_spin_lock_nested(&parent->lock, SINGLE_DEPTH_NESTING);
+
+ child->parent = parent;
+ child->childmask = BIT(parent->num_children++);
+
+ raw_spin_unlock(&parent->lock);
+ raw_spin_unlock_irq(&child->lock);
+
+ /*
+ * To prevent inconsistent states, active children need to be active in
+ * the new parent as well. Inactive children are already marked inactive
+ * in the parent group.
+ */
+ childstate.state = atomic_read(&child->migr_state);
+ if (childstate.migrator != TMIGR_NONE) {
+ struct tmigr_walk data;
+
+ data.childmask = child->childmask;
+ data.groupstate.state = atomic_read(&parent->migr_state);
+
+ /*
+ * There is only one new level per time. When connecting the
+ * child and the parent and set the child active when the parent
+ * is inactive, the parent needs to be the uppermost
+ * level. Otherwise there went something wrong!
+ */
+ WARN_ON(!tmigr_active_up(parent, child, &data) && parent->parent);
+ }
+}
+
+static int tmigr_setup_groups(unsigned int cpu, unsigned int node)
+{
+ struct tmigr_group *group, *child, **stack;
+ int top = 0, err = 0, i = 0;
+ struct list_head *lvllist;
+
+ stack = kcalloc(tmigr_hierarchy_levels, sizeof(*stack), GFP_KERNEL);
+ if (!stack)
+ return -ENOMEM;
+
+ do {
+ group = tmigr_get_group(cpu, node, i);
+ if (IS_ERR(group)) {
+ err = PTR_ERR(group);
+ break;
+ }
+
+ top = i;
+ stack[i++] = group;
+
+ /*
+ * When booting only less CPUs of a system than CPUs are
+ * available, not all calculated hierarchy levels are required.
+ *
+ * The loop is aborted as soon as the highest level, which might
+ * be different from tmigr_hierarchy_levels, contains only a
+ * single group.
+ */
+ if (group->parent || i == tmigr_hierarchy_levels ||
+ (list_empty(&tmigr_level_list[i]) &&
+ list_is_singular(&tmigr_level_list[i - 1])))
+ break;
+
+ } while (i < tmigr_hierarchy_levels);
+
+ do {
+ group = stack[--i];
+
+ if (err < 0) {
+ list_del(&group->list);
+ kfree(group);
+ continue;
+ }
+
+ WARN_ON_ONCE(i != group->level);
+
+ /*
+ * Update tmc -> group / child -> group connection
+ */
+ if (i == 0) {
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+
+ raw_spin_lock_irq(&group->lock);
+
+ tmc->tmgroup = group;
+ tmc->childmask = BIT(group->num_children++);
+
+ raw_spin_unlock_irq(&group->lock);
+
+ /* There are no children that need to be connected */
+ continue;
+ } else {
+ child = stack[i - 1];
+ tmigr_connect_child_parent(child, group);
+ }
+
+ /* check if uppermost level was newly created */
+ if (top != i)
+ continue;
+
+ WARN_ON_ONCE(top == 0);
+
+ lvllist = &tmigr_level_list[top];
+ if (group->num_children == 1 && list_is_singular(lvllist)) {
+ lvllist = &tmigr_level_list[top - 1];
+ list_for_each_entry(child, lvllist, list) {
+ if (child->parent)
+ continue;
+
+ tmigr_connect_child_parent(child, group);
+ }
+ }
+ } while (i > 0);
+
+ kfree(stack);
+
+ return err;
+}
+
+static int tmigr_add_cpu(unsigned int cpu)
+{
+ int node = cpu_to_node(cpu);
+ int ret;
+
+ mutex_lock(&tmigr_mutex);
+ ret = tmigr_setup_groups(cpu, node);
+ mutex_unlock(&tmigr_mutex);
+
+ return ret;
+}
+
+static int tmigr_cpu_online(unsigned int cpu)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+ unsigned int ret;
+
+ /* First online attempt? Initialize CPU data */
+ if (!tmc->tmgroup) {
+ raw_spin_lock_init(&tmc->lock);
+
+ ret = tmigr_add_cpu(cpu);
+ if (ret < 0)
+ return ret;
+
+ if (tmc->childmask == 0)
+ return -EINVAL;
+
+ timerqueue_init(&tmc->cpuevt.nextevt);
+ tmc->cpuevt.nextevt.expires = KTIME_MAX;
+ tmc->cpuevt.ignore = true;
+ tmc->cpuevt.cpu = cpu;
+
+ tmc->remote = false;
+ tmc->idle = false;
+ WRITE_ONCE(tmc->wakeup, KTIME_MAX);
+ }
+ raw_spin_lock_irq(&tmc->lock);
+ __tmigr_cpu_activate(tmc);
+ tmc->online = true;
+ raw_spin_unlock_irq(&tmc->lock);
+ return 0;
+}
+
+static int tmigr_cpu_offline(unsigned int cpu)
+{
+ struct tmigr_cpu *tmc = this_cpu_ptr(&tmigr_cpu);
+
+ raw_spin_lock_irq(&tmc->lock);
+ tmc->online = false;
+ WRITE_ONCE(tmc->wakeup, KTIME_MAX);
+
+ /*
+ * CPU has to handle the local events on his own, when on the way to
+ * offline; Therefore nextevt value is set to KTIME_MAX
+ */
+ __tmigr_cpu_deactivate(tmc, KTIME_MAX);
+ raw_spin_unlock_irq(&tmc->lock);
+
+ return 0;
+}
+
+static int __init tmigr_init(void)
+{
+ unsigned int cpulvl, nodelvl, cpus_per_node, i;
+ unsigned int nnodes = num_possible_nodes();
+ unsigned int ncpus = num_possible_cpus();
+ int ret = -ENOMEM;
+
+ /* Nothing to do if running on UP */
+ if (ncpus == 1)
+ return 0;
+
+ /*
+ * Calculate the required hierarchy levels. Unfortunately there is no
+ * reliable information available, unless all possible CPUs have been
+ * brought up and all numa nodes are populated.
+ *
+ * Estimate the number of levels with the number of possible nodes and
+ * the number of possible CPUs. Assume CPUs are spread evenly across
+ * nodes. We cannot rely on cpumask_of_node() because there only already
+ * online CPUs are considered.
+ */
+ cpus_per_node = DIV_ROUND_UP(ncpus, nnodes);
+
+ /* Calc the hierarchy levels required to hold the CPUs of a node */
+ cpulvl = DIV_ROUND_UP(order_base_2(cpus_per_node),
+ ilog2(TMIGR_CHILDREN_PER_GROUP));
+
+ /* Calculate the extra levels to connect all nodes */
+ nodelvl = DIV_ROUND_UP(order_base_2(nnodes),
+ ilog2(TMIGR_CHILDREN_PER_GROUP));
+
+ tmigr_hierarchy_levels = cpulvl + nodelvl;
+
+ /*
+ * If a numa node spawns more than one CPU level group then the next
+ * level(s) of the hierarchy contains groups which handle all CPU groups
+ * of the same numa node. The level above goes across numa nodes. Store
+ * this information for the setup code to decide when node matching is
+ * not longer required.
+ */
+ tmigr_crossnode_level = cpulvl;
+
+ tmigr_level_list = kcalloc(tmigr_hierarchy_levels, sizeof(struct list_head), GFP_KERNEL);
+ if (!tmigr_level_list)
+ goto err;
+
+ for (i = 0; i < tmigr_hierarchy_levels; i++)
+ INIT_LIST_HEAD(&tmigr_level_list[i]);
+
+ pr_info("Timer migration: %d hierarchy levels; %d children per group;"
+ " %d crossnode level\n",
+ tmigr_hierarchy_levels, TMIGR_CHILDREN_PER_GROUP,
+ tmigr_crossnode_level);
+
+ ret = cpuhp_setup_state(CPUHP_AP_TMIGR_ONLINE, "tmigr:online",
+ tmigr_cpu_online, tmigr_cpu_offline);
+ if (ret)
+ goto err;
+
+ static_branch_enable(&tmigr_enabled);
+
+ return 0;
+
+err:
+ pr_err("Timer migration setup failed\n");
+ return ret;
+}
+late_initcall(tmigr_init);
diff --git a/kernel/time/timer_migration.h b/kernel/time/timer_migration.h
new file mode 100644
index 000000000000..01c08a5dbb49
--- /dev/null
+++ b/kernel/time/timer_migration.h
@@ -0,0 +1,144 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef _KERNEL_TIME_MIGRATION_H
+#define _KERNEL_TIME_MIGRATION_H
+
+/* Per group capacity. Must be a power of 2! */
+#define TMIGR_CHILDREN_PER_GROUP 8
+
+/**
+ * struct tmigr_event - a timer event associated to a CPU
+ * @nextevt: The node to enqueue an event in the parent group queue
+ * @cpu: The CPU to which this event belongs
+ * @ignore: Hint whether the event could be ignored; it is set when
+ * CPU or group is active;
+ */
+struct tmigr_event {
+ struct timerqueue_node nextevt;
+ unsigned int cpu;
+ bool ignore;
+};
+
+/**
+ * struct tmigr_group - timer migration hierarchy group
+ * @lock: Lock protecting the event information and group hierarchy
+ * information during setup
+ * @migr_state: State of group (see union tmigr_state)
+ * @parent: Pointer to parent group
+ * @groupevt: Next event of group which is only used when group is
+ * !active. The group event is then queued into parent
+ * timer queue.
+ * Ignore bit of @groupevt is set when group is active.
+ * @next_expiry: Base monotonic expiry time of next event of group;
+ * Used for racy lockless check whether remote expiry is
+ * required; it is always reliable
+ * @events: Timer queue for child events queued in the group
+ * @childmask: childmask of group in parent group; is set during
+ * setup and will never change; could be read lockless
+ * @level: Hierarchy level of group; Required during setup and
+ * when DEBUG is defined
+ * @list: List head that is added to per level
+ * tmigr_level_list; is required during setup when a
+ * new group needs to be connected to the existing
+ * hierarchy groups
+ * @numa_node: Is set to numa node when level < tmigr_crossnode_level;
+ * otherwise it is set to NUMA_NO_NODE; Required for
+ * setup only to make sure CPUs and groups are per
+ * numa node as long as level < tmigr_crossnode_level
+ * @num_children: Counter of group children to make sure group is only filled
+ * with TMIGR_CHILDREN_PER_GROUP; Required for setup only
+ */
+struct tmigr_group {
+ raw_spinlock_t lock;
+ atomic_t migr_state;
+ struct tmigr_group *parent;
+ struct tmigr_event groupevt;
+ u64 next_expiry;
+ struct timerqueue_head events;
+ u8 childmask;
+ unsigned int level;
+ struct list_head list;
+ int numa_node;
+ unsigned int num_children;
+};
+
+/**
+ * struct tmigr_cpu - timer migration per CPU group
+ * @lock: Lock protecting tmigr_cpu group information
+ * @online: Indicates whether CPU is online; In deactivate path it
+ * is required to know whether the migrator in top level
+ * group is on the way to go offline when a timer is
+ * pending. Then another online CPU needs to be rescheduled
+ * to make sure timers are handled properly; Furthermore
+ * the information is required in CPU hotplug path as CPU is
+ * able to go idle before timer migration hierarchy hotplug
+ * AP is reached. During this phase CPU has to handle
+ * global timers by its own and does not act as migrator.
+ * @idle: Indicates whether CPU is idle in timer migration
+ * hierarchy
+ * @remote: Is set when timers of CPU are expired remote
+ * @wakeup_recalc: Indicates, whether a recalculation of @wakeup value is
+ * required. It is only used when CPU is marked idle in
+ * timer migration hierarchy.
+ * @tmgroup: Pointer to parent group
+ * @childmask: childmask of tmigr_cpu in parent group
+ * @wakeup: Stores the first timer when the timer migration
+ * hierarchy is completely idle and remote expiry was done;
+ * is returned to timer code when tmigr_cpu_deactive() is
+ * called and group is idle; it is only valid, when
+ * @wakeup_recalc is not set;
+ * @cpuevt: CPU event which could be queued into parent group
+ */
+struct tmigr_cpu {
+ raw_spinlock_t lock;
+ bool online;
+ bool idle;
+ bool remote;
+ bool wakeup_recalc;
+ struct tmigr_group *tmgroup;
+ u8 childmask;
+ u64 wakeup;
+ struct tmigr_event cpuevt;
+};
+
+/**
+ * union tmigr_state - state of tmigr_group
+ * @state: Combined version of the state - only used for atomic
+ * read/cmpxchg function
+ * @struct: Split version of the state - only use the struct members to
+ * update information to stay independent of endianness
+ */
+union tmigr_state {
+ u32 state;
+ /**
+ * struct - split state of tmigr_group
+ * @active: Contains each childmask bit of active children
+ * @migrator: Contains childmask of child which is migrator
+ * @seq: Sequence counter needs to be increased when update
+ * to the tmigr_state is done. It prevents a race when
+ * updates in child groups are propagated in changed
+ * order. Detailed information about the scenario is
+ * given in documentation at the begin of
+ * timer_migration.c.
+ */
+ struct {
+ u8 active;
+ u8 migrator;
+ u16 seq;
+ } __packed;
+};
+
+#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
+extern void tmigr_handle_remote(void);
+extern int tmigr_requires_handle_remote(void);
+extern void tmigr_cpu_activate(void);
+extern u64 tmigr_cpu_deactivate(u64 nextevt);
+extern u64 tmigr_quick_check(void);
+#else
+static inline void tmigr_handle_remote(void) { }
+static inline int tmigr_requires_handle_remote(void) { return 0; }
+static inline void tmigr_cpu_activate(void) { }
+static inline u64 tmigr_cpu_deactivate(u64 nextevt) { return KTIME_MAX; }
+static inline u64 tmigr_quick_check(void) { return KTIME_MAX; };
+#endif
+
+#endif
--
2.39.2
The logic for raising a softirq the way it is implemented right now, is
readable for two timer bases. When increasing numbers of timer bases, code
gets harder to read. With the introduction of the timer migration
hierarchy, there will be three timer bases.
Therefore ease the code. No functional change.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
v5: New patch to decrease patch size of follow up patches
---
kernel/time/timer.c | 14 ++++++--------
1 file changed, 6 insertions(+), 8 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 853a8ee48ffc..1aafa0edc74c 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -2123,16 +2123,14 @@ static void run_local_timers(void)
struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
hrtimer_run_queues();
- /* Raise the softirq only if required. */
- if (time_before(jiffies, base->next_expiry)) {
- if (!IS_ENABLED(CONFIG_NO_HZ_COMMON))
- return;
- /* CPU is awake, so check the deferrable base. */
- base++;
- if (time_before(jiffies, base->next_expiry))
+
+ for (int i = 0; i < NR_BASES; i++, base++) {
+ /* Raise the softirq only if required. */
+ if (time_after_eq(jiffies, base->next_expiry)) {
+ raise_softirq(TIMER_SOFTIRQ);
return;
+ }
}
- raise_softirq(TIMER_SOFTIRQ);
}
/*
--
2.39.2
From: "Richard Cochran (linutronix GmbH)" <[email protected]>
Move the locking out from __run_timers() to the call sites, so the
protected section can be extended at the call site. Preparatory patch for
changing the NOHZ timer placement to a pull at expiry time model.
No functional change.
Signed-off-by: Richard Cochran (linutronix GmbH) <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
kernel/time/timer.c | 31 +++++++++++++++++++++----------
1 file changed, 21 insertions(+), 10 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index ae4b6f62b082..8893f5dd1d66 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -2230,11 +2230,7 @@ static inline void __run_timers(struct timer_base *base)
struct hlist_head heads[LVL_DEPTH];
int levels;
- if (time_before(jiffies, base->next_expiry))
- return;
-
- timer_base_lock_expiry(base);
- raw_spin_lock_irq(&base->lock);
+ lockdep_assert_held(&base->lock);
while (time_after_eq(jiffies, base->clk) &&
time_after_eq(jiffies, base->next_expiry)) {
@@ -2254,21 +2250,36 @@ static inline void __run_timers(struct timer_base *base)
while (levels--)
expire_timers(base, heads + levels);
}
+}
+
+static void __run_timer_base(struct timer_base *base)
+{
+ if (time_before(jiffies, base->next_expiry))
+ return;
+
+ timer_base_lock_expiry(base);
+ raw_spin_lock_irq(&base->lock);
+ __run_timers(base);
raw_spin_unlock_irq(&base->lock);
timer_base_unlock_expiry(base);
}
+static void run_timer_base(int index)
+{
+ struct timer_base *base = this_cpu_ptr(&timer_bases[index]);
+
+ __run_timer_base(base);
+}
+
/*
* This function runs timers and the timer-tq in bottom half context.
*/
static __latent_entropy void run_timer_softirq(struct softirq_action *h)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_LOCAL]);
-
- __run_timers(base);
+ run_timer_base(BASE_LOCAL);
if (IS_ENABLED(CONFIG_NO_HZ_COMMON)) {
- __run_timers(this_cpu_ptr(&timer_bases[BASE_GLOBAL]));
- __run_timers(this_cpu_ptr(&timer_bases[BASE_DEF]));
+ run_timer_base(BASE_GLOBAL);
+ run_timer_base(BASE_DEF);
}
}
--
2.39.2
Separate the storage space for pinned timers. Deferrable timers (doesn't
matter if pinned or non pinned) are still enqueued into their own base.
This is preparatory work for changing the NOHZ timer placement from a push
at enqueue time to a pull at expiry time model.
Originally-by: Richard Cochran (linutronix GmbH) <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
v6:
- Drop set TIMER_PINNED flag in add_timer_on() and drop related
warning. add_timer_on() fix is splitted into a separate
patch. Therefore also drop "Reviewed-by" of Frederic Weisbecker
v5:
- Add WARN_ONCE() in add_timer_on()
- Decrease patch size by splitting into three patches (this patch and the
two before)
v4:
- split out logic to forward base clock into a helper function
forward_base_clk() (Frederic)
- ease the code in run_local_timers() and timer_clear_idle() (Frederic)
---
kernel/time/timer.c | 106 ++++++++++++++++++++++++++++++++------------
1 file changed, 77 insertions(+), 29 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 7468a805d302..1cd214c2f3f1 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -187,12 +187,18 @@ EXPORT_SYMBOL(jiffies_64);
#define WHEEL_SIZE (LVL_SIZE * LVL_DEPTH)
#ifdef CONFIG_NO_HZ_COMMON
-# define NR_BASES 2
-# define BASE_STD 0
-# define BASE_DEF 1
+/*
+ * If multiple bases need to be locked, use the base ordering for lock
+ * nesting, i.e. lowest number first.
+ */
+# define NR_BASES 3
+# define BASE_LOCAL 0
+# define BASE_GLOBAL 1
+# define BASE_DEF 2
#else
# define NR_BASES 1
-# define BASE_STD 0
+# define BASE_LOCAL 0
+# define BASE_GLOBAL 0
# define BASE_DEF 0
#endif
@@ -899,7 +905,10 @@ static int detach_if_pending(struct timer_list *timer, struct timer_base *base,
static inline struct timer_base *get_timer_cpu_base(u32 tflags, u32 cpu)
{
- struct timer_base *base = per_cpu_ptr(&timer_bases[BASE_STD], cpu);
+ int index = tflags & TIMER_PINNED ? BASE_LOCAL : BASE_GLOBAL;
+ struct timer_base *base;
+
+ base = per_cpu_ptr(&timer_bases[index], cpu);
/*
* If the timer is deferrable and NO_HZ_COMMON is set then we need
@@ -912,7 +921,10 @@ static inline struct timer_base *get_timer_cpu_base(u32 tflags, u32 cpu)
static inline struct timer_base *get_timer_this_cpu_base(u32 tflags)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
+ int index = tflags & TIMER_PINNED ? BASE_LOCAL : BASE_GLOBAL;
+ struct timer_base *base;
+
+ base = this_cpu_ptr(&timer_bases[index]);
/*
* If the timer is deferrable and NO_HZ_COMMON is set then we need
@@ -1960,13 +1972,31 @@ static unsigned long next_timer_interrupt(struct timer_base *base)
}
static inline unsigned long __get_next_timer_interrupt(unsigned long basej,
- struct timer_base *base)
+ struct timer_base *base_local,
+ struct timer_base *base_global)
{
- unsigned long nextevt;
+ unsigned long nextevt, nextevt_local, nextevt_global;
+ bool local_first;
- nextevt = next_timer_interrupt(base);
+ nextevt_local = next_timer_interrupt(base_local);
+ nextevt_global = next_timer_interrupt(base_global);
- if (base->timers_pending) {
+ /*
+ * Check whether the local event is expiring before or at the same
+ * time as the global event.
+ *
+ * Note, that nextevt_global and nextevt_local might be based on
+ * different base->clk values. So it's not guaranteed that
+ * comparing with empty bases results in a correct local_first.
+ */
+ if (base_local->timers_pending && base_global->timers_pending)
+ local_first = time_before_eq(nextevt_local, nextevt_global);
+ else
+ local_first = base_local->timers_pending;
+
+ nextevt = local_first ? nextevt_local : nextevt_global;
+
+ if (base_local->timers_pending || base_global->timers_pending) {
/* If we missed a tick already, force 0 delta */
if (time_before(nextevt, basej))
nextevt = basej;
@@ -1985,7 +2015,7 @@ static inline unsigned long __get_next_timer_interrupt(unsigned long basej,
*/
u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
+ struct timer_base *base_local, *base_global;
u64 expires = KTIME_MAX;
unsigned long nextevt;
@@ -1996,9 +2026,16 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
if (cpu_is_offline(smp_processor_id()))
return expires;
- raw_spin_lock(&base->lock);
- nextevt = __get_next_timer_interrupt(basej, base);
- raw_spin_unlock(&base->lock);
+ base_local = this_cpu_ptr(&timer_bases[BASE_LOCAL]);
+ base_global = this_cpu_ptr(&timer_bases[BASE_GLOBAL]);
+
+ raw_spin_lock(&base_local->lock);
+ raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
+
+ nextevt = __get_next_timer_interrupt(basej, base_local, base_global);
+
+ raw_spin_unlock(&base_global->lock);
+ raw_spin_unlock(&base_local->lock);
expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
@@ -2018,7 +2055,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
*/
u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
+ struct timer_base *base_local, *base_global;
unsigned long nextevt;
/*
@@ -2030,25 +2067,35 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
return KTIME_MAX;
}
- raw_spin_lock(&base->lock);
- nextevt = __get_next_timer_interrupt(basej, base);
+ base_local = this_cpu_ptr(&timer_bases[BASE_LOCAL]);
+ base_global = this_cpu_ptr(&timer_bases[BASE_GLOBAL]);
+
+ raw_spin_lock(&base_local->lock);
+ raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
+
+ nextevt = __get_next_timer_interrupt(basej, base_local, base_global);
/*
* We have a fresh next event. Check whether we can forward the
* base.
*/
- __forward_timer_base(base, basej);
+ __forward_timer_base(base_local, basej);
+ __forward_timer_base(base_global, basej);
/*
- * Base is idle if the next event is more than a tick away. Also
+ * Bases are idle if the next event is more than a tick away. Also
* the tick is stopped so any added timer must forward the base clk
* itself to keep granularity small. This idle logic is only
- * maintained for the BASE_STD base, deferrable timers may still
- * see large granularity skew (by design).
+ * maintained for the BASE_LOCAL and BASE_GLOBAL base, deferrable
+ * timers may still see large granularity skew (by design).
*/
- base->is_idle = *idle = time_after(nextevt, basej + 1);
+ *idle = time_after(nextevt, basej + 1);
+
+ /* We need to mark both bases in sync */
+ base_local->is_idle = base_global->is_idle = *idle;
- raw_spin_unlock(&base->lock);
+ raw_spin_unlock(&base_global->lock);
+ raw_spin_unlock(&base_local->lock);
return basem + (u64)(nextevt - basej) * TICK_NSEC;
}
@@ -2060,15 +2107,14 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
*/
void timer_clear_idle(void)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
-
/*
* We do this unlocked. The worst outcome is a remote enqueue sending
* a pointless IPI, but taking the lock would just make the window for
* sending the IPI a few instructions smaller for the cost of taking
* the lock in the exit from idle path.
*/
- base->is_idle = false;
+ __this_cpu_write(timer_bases[BASE_LOCAL].is_idle, false);
+ __this_cpu_write(timer_bases[BASE_GLOBAL].is_idle, false);
}
#endif
@@ -2114,11 +2160,13 @@ static inline void __run_timers(struct timer_base *base)
*/
static __latent_entropy void run_timer_softirq(struct softirq_action *h)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
+ struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_LOCAL]);
__run_timers(base);
- if (IS_ENABLED(CONFIG_NO_HZ_COMMON))
+ if (IS_ENABLED(CONFIG_NO_HZ_COMMON)) {
+ __run_timers(this_cpu_ptr(&timer_bases[BASE_GLOBAL]));
__run_timers(this_cpu_ptr(&timer_bases[BASE_DEF]));
+ }
}
/*
@@ -2126,7 +2174,7 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h)
*/
static void run_local_timers(void)
{
- struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
+ struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_LOCAL]);
hrtimer_run_queues();
--
2.39.2
Due to the conversion of the NOHZ timer placement to a pull at expiry
time model, the per CPU timer bases with non pinned timers are no
longer handled only by the local CPU. In case a remote CPU already
expires the non pinned timers base of the local cpu, nothing more
needs to be done by the local CPU. A check at the begin of the expire
timers routine is required, because timer base lock is dropped before
executing the timer callback function.
This is a preparatory work, but has no functional impact right now.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
v6: Drop double negation
---
kernel/time/timer.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 8893f5dd1d66..7f06553e44de 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -2232,6 +2232,9 @@ static inline void __run_timers(struct timer_base *base)
lockdep_assert_held(&base->lock);
+ if (base->running_timer)
+ return;
+
while (time_after_eq(jiffies, base->clk) &&
time_after_eq(jiffies, base->next_expiry)) {
levels = collect_expired_timers(base, heads);
--
2.39.2
To prepare for the conversion of the NOHZ timer placement to a pull at
expiry time model it's required to have functionality available getting the
next timer interrupt on a remote CPU.
Locking of the timer bases and getting the information for the next timer
interrupt functionality is split into separate functions. This is required
to be compliant with lock ordering when the new model is in place.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
v8:
- Update comment
v7:
- Move functions into CONFIG_SMP && CONFIG_NO_HZ_COMMON section
- change lock, fetch functions to be unconditional
- split out unlock function into a separate function
v6:
- introduce timer_lock_remote_bases() to fix race
---
kernel/time/tick-internal.h | 10 +++++
kernel/time/timer.c | 76 ++++++++++++++++++++++++++++++++++---
2 files changed, 81 insertions(+), 5 deletions(-)
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index b035606a6f5e..206010ae2a53 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -8,6 +8,11 @@
#include "timekeeping.h"
#include "tick-sched.h"
+struct timer_events {
+ u64 local;
+ u64 global;
+};
+
#ifdef CONFIG_GENERIC_CLOCKEVENTS
# define TICK_DO_TIMER_NONE -1
@@ -155,6 +160,11 @@ extern unsigned long tick_nohz_active;
extern void timers_update_nohz(void);
# ifdef CONFIG_SMP
extern struct static_key_false timers_migration_enabled;
+extern void fetch_next_timer_interrupt_remote(unsigned long basej, u64 basem,
+ struct timer_events *tevt,
+ unsigned int cpu);
+extern void timer_lock_remote_bases(unsigned int cpu);
+extern void timer_unlock_remote_bases(unsigned int cpu);
# endif
#else /* CONFIG_NO_HZ_COMMON */
static inline void timers_update_nohz(void) { }
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index c3061b28214e..ae4b6f62b082 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -221,11 +221,6 @@ struct timer_base {
static DEFINE_PER_CPU(struct timer_base, timer_bases[NR_BASES]);
-struct timer_events {
- u64 local;
- u64 global;
-};
-
#ifdef CONFIG_NO_HZ_COMMON
static DEFINE_STATIC_KEY_FALSE(timers_nohz_active);
@@ -2016,6 +2011,77 @@ static unsigned long fetch_next_timer_interrupt(unsigned long basej, u64 basem,
return local_first ? nextevt_local : nextevt_global;
}
+# ifdef CONFIG_SMP
+/**
+ * fetch_next_timer_interrupt_remote
+ * @basej: base time jiffies
+ * @basem: base time clock monotonic
+ * @tevt: Pointer to the storage for the expiry values
+ * @cpu: Remote CPU
+ *
+ * Stores the next pending local and global timer expiry values in the
+ * struct pointed to by @tevt. If a queue is empty the corresponding
+ * field is set to KTIME_MAX. If local event expires before global
+ * event, global event is set to KTIME_MAX as well.
+ *
+ * Caller needs to make sure timer base locks are held (use
+ * timer_lock_remote_bases() for this purpose).
+ */
+void fetch_next_timer_interrupt_remote(unsigned long basej, u64 basem,
+ struct timer_events *tevt,
+ unsigned int cpu)
+{
+ struct timer_base *base_local, *base_global;
+
+ /* Preset local / global events */
+ tevt->local = tevt->global = KTIME_MAX;
+
+ base_local = per_cpu_ptr(&timer_bases[BASE_LOCAL], cpu);
+ base_global = per_cpu_ptr(&timer_bases[BASE_GLOBAL], cpu);
+
+ lockdep_assert_held(&base_local->lock);
+ lockdep_assert_held(&base_global->lock);
+
+ fetch_next_timer_interrupt(basej, basem, base_local, base_global, tevt);
+}
+
+/**
+ * timer_unlock_remote_bases - unlock timer bases of cpu
+ * @cpu: Remote CPU
+ *
+ * Unlocks the remote timer bases.
+ */
+void timer_unlock_remote_bases(unsigned int cpu)
+{
+ struct timer_base *base_local, *base_global;
+
+ base_local = per_cpu_ptr(&timer_bases[BASE_LOCAL], cpu);
+ base_global = per_cpu_ptr(&timer_bases[BASE_GLOBAL], cpu);
+
+ raw_spin_unlock(&base_global->lock);
+ raw_spin_unlock(&base_local->lock);
+}
+
+/**
+ * timer_lock_remote_bases - lock timer bases of cpu
+ * @cpu: Remote CPU
+ *
+ * Locks the remote timer bases.
+ */
+void timer_lock_remote_bases(unsigned int cpu)
+{
+ struct timer_base *base_local, *base_global;
+
+ base_local = per_cpu_ptr(&timer_bases[BASE_LOCAL], cpu);
+ base_global = per_cpu_ptr(&timer_bases[BASE_GLOBAL], cpu);
+
+ lockdep_assert_irqs_disabled();
+
+ raw_spin_lock(&base_local->lock);
+ raw_spin_lock_nested(&base_global->lock, SINGLE_DEPTH_NESTING);
+}
+# endif /* CONFIG_SMP */
+
static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64 basem,
struct timer_base *base_local,
struct timer_base *base_global,
--
2.39.2
The timer pull logic needs proper debugging aids. Add tracepoints so the
hierarchical idle machinery can be diagnosed.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
v8: Add wakeup value to tracepoints
---
include/trace/events/timer_migration.h | 283 +++++++++++++++++++++++++
kernel/time/timer_migration.c | 24 +++
2 files changed, 307 insertions(+)
create mode 100644 include/trace/events/timer_migration.h
diff --git a/include/trace/events/timer_migration.h b/include/trace/events/timer_migration.h
new file mode 100644
index 000000000000..b8232c1a1ebf
--- /dev/null
+++ b/include/trace/events/timer_migration.h
@@ -0,0 +1,283 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM timer_migration
+
+#if !defined(_TRACE_TIMER_MIGRATION_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_TIMER_MIGRATION_H
+
+#include <linux/tracepoint.h>
+
+/* Group events */
+TRACE_EVENT(tmigr_group_set,
+
+ TP_PROTO(struct tmigr_group *group),
+
+ TP_ARGS(group),
+
+ TP_STRUCT__entry(
+ __field( void *, group )
+ __field( unsigned int, lvl )
+ __field( unsigned int, numa_node )
+ ),
+
+ TP_fast_assign(
+ __entry->group = group;
+ __entry->lvl = group->level;
+ __entry->numa_node = group->numa_node;
+ ),
+
+ TP_printk("group=%p lvl=%d numa=%d",
+ __entry->group, __entry->lvl, __entry->numa_node)
+);
+
+TRACE_EVENT(tmigr_connect_child_parent,
+
+ TP_PROTO(struct tmigr_group *child),
+
+ TP_ARGS(child),
+
+ TP_STRUCT__entry(
+ __field( void *, child )
+ __field( void *, parent )
+ __field( unsigned int, lvl )
+ __field( unsigned int, numa_node )
+ __field( unsigned int, num_children )
+ __field( u32, childmask )
+ ),
+
+ TP_fast_assign(
+ __entry->child = child;
+ __entry->parent = child->parent;
+ __entry->lvl = child->parent->level;
+ __entry->numa_node = child->parent->numa_node;
+ __entry->numa_node = child->parent->num_children;
+ __entry->childmask = child->childmask;
+ ),
+
+ TP_printk("group=%p childmask=%0x parent=%p lvl=%d numa=%d num_children=%d",
+ __entry->child, __entry->childmask, __entry->parent,
+ __entry->lvl, __entry->numa_node, __entry->num_children)
+);
+
+TRACE_EVENT(tmigr_connect_cpu_parent,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc),
+
+ TP_STRUCT__entry(
+ __field( void *, parent )
+ __field( unsigned int, cpu )
+ __field( unsigned int, lvl )
+ __field( unsigned int, numa_node )
+ __field( unsigned int, num_children )
+ __field( u32, childmask )
+ ),
+
+ TP_fast_assign(
+ __entry->parent = tmc->tmgroup;
+ __entry->cpu = tmc->cpuevt.cpu;
+ __entry->lvl = tmc->tmgroup->level;
+ __entry->numa_node = tmc->tmgroup->numa_node;
+ __entry->numa_node = tmc->tmgroup->num_children;
+ __entry->childmask = tmc->childmask;
+ ),
+
+ TP_printk("cpu=%d childmask=%0x parent=%p lvl=%d numa=%d num_children=%d",
+ __entry->cpu, __entry->childmask, __entry->parent,
+ __entry->lvl, __entry->numa_node, __entry->num_children)
+);
+
+DECLARE_EVENT_CLASS(tmigr_group_and_cpu,
+
+ TP_PROTO(struct tmigr_group *group, union tmigr_state state, u32 childmask),
+
+ TP_ARGS(group, state, childmask),
+
+ TP_STRUCT__entry(
+ __field( void *, group )
+ __field( void *, parent )
+ __field( unsigned int, lvl )
+ __field( unsigned int, numa_node )
+ __field( u8, active )
+ __field( u8, migrator )
+ __field( u32, childmask )
+ ),
+
+ TP_fast_assign(
+ __entry->group = group;
+ __entry->parent = group->parent;
+ __entry->lvl = group->level;
+ __entry->numa_node = group->numa_node;
+ __entry->active = state.active;
+ __entry->migrator = state.migrator;
+ __entry->childmask = childmask;
+ ),
+
+ TP_printk("group=%p lvl=%d numa=%d active=%0x migrator=%0x "
+ "parent=%p childmask=%0x",
+ __entry->group, __entry->lvl, __entry->numa_node,
+ __entry->active, __entry->migrator,
+ __entry->parent, __entry->childmask)
+);
+
+DEFINE_EVENT(tmigr_group_and_cpu, tmigr_group_set_cpu_inactive,
+
+ TP_PROTO(struct tmigr_group *group, union tmigr_state state, u32 childmask),
+
+ TP_ARGS(group, state, childmask)
+);
+
+DEFINE_EVENT(tmigr_group_and_cpu, tmigr_group_set_cpu_active,
+
+ TP_PROTO(struct tmigr_group *group, union tmigr_state state, u32 childmask),
+
+ TP_ARGS(group, state, childmask)
+);
+
+/* CPU events*/
+DECLARE_EVENT_CLASS(tmigr_cpugroup,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc),
+
+ TP_STRUCT__entry(
+ __field( void *, parent)
+ __field( unsigned int, cpu)
+ __field( u64, wakeup)
+ ),
+
+ TP_fast_assign(
+ __entry->cpu = tmc->cpuevt.cpu;
+ __entry->parent = tmc->tmgroup;
+ __entry->wakeup = tmc->wakeup;
+ ),
+
+ TP_printk("cpu=%d parent=%p wakeup=%llu", __entry->cpu, __entry->parent, __entry->wakeup)
+);
+
+DEFINE_EVENT(tmigr_cpugroup, tmigr_cpu_new_timer,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc)
+);
+
+DEFINE_EVENT(tmigr_cpugroup, tmigr_cpu_active,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc)
+);
+
+DEFINE_EVENT(tmigr_cpugroup, tmigr_cpu_online,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc)
+);
+
+DEFINE_EVENT(tmigr_cpugroup, tmigr_cpu_offline,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc)
+);
+
+DEFINE_EVENT(tmigr_cpugroup, tmigr_handle_remote_cpu,
+
+ TP_PROTO(struct tmigr_cpu *tmc),
+
+ TP_ARGS(tmc)
+);
+
+TRACE_EVENT(tmigr_cpu_idle,
+
+ TP_PROTO(struct tmigr_cpu *tmc, u64 nextevt),
+
+ TP_ARGS(tmc, nextevt),
+
+ TP_STRUCT__entry(
+ __field( void *, parent)
+ __field( unsigned int, cpu)
+ __field( u64, nextevt)
+ __field( u64, wakeup)
+ ),
+
+ TP_fast_assign(
+ __entry->cpu = tmc->cpuevt.cpu;
+ __entry->parent = tmc->tmgroup;
+ __entry->nextevt = nextevt;
+ __entry->wakeup = tmc->wakeup;
+ ),
+
+ TP_printk("cpu=%d parent=%p nextevt=%llu wakeup=%llu",
+ __entry->cpu, __entry->parent, __entry->nextevt, __entry->wakeup)
+);
+
+TRACE_EVENT(tmigr_update_events,
+
+ TP_PROTO(struct tmigr_group *child, struct tmigr_group *group,
+ union tmigr_state childstate, union tmigr_state groupstate,
+ u64 nextevt),
+
+ TP_ARGS(child, group, childstate, groupstate, nextevt),
+
+ TP_STRUCT__entry(
+ __field( void *, child )
+ __field( void *, group )
+ __field( u64, nextevt )
+ __field( u64, group_next_expiry )
+ __field( unsigned int, group_lvl )
+ __field( u8, child_active )
+ __field( u8, group_active )
+ __field( unsigned int, child_evtcpu )
+ __field( u64, child_evt_expiry )
+ ),
+
+ TP_fast_assign(
+ __entry->child = child;
+ __entry->group = group;
+ __entry->nextevt = nextevt;
+ __entry->group_next_expiry = group->next_expiry;
+ __entry->group_lvl = group->level;
+ __entry->child_active = childstate.active;
+ __entry->group_active = groupstate.active;
+ __entry->child_evtcpu = child ? child->groupevt.cpu : 0;
+ __entry->child_evt_expiry = child ? child->groupevt.nextevt.expires : 0;
+ ),
+
+ TP_printk("child=%p group=%p group_lvl=%d child_active=%0x group_active=%0x "
+ "nextevt=%llu next_expiry=%llu child_evt_expiry=%llu child_evtcpu=%d",
+ __entry->child, __entry->group, __entry->group_lvl, __entry->child_active,
+ __entry->group_active,
+ __entry->nextevt, __entry->group_next_expiry, __entry->child_evt_expiry,
+ __entry->child_evtcpu)
+);
+
+TRACE_EVENT(tmigr_handle_remote,
+
+ TP_PROTO(struct tmigr_group *group),
+
+ TP_ARGS(group),
+
+ TP_STRUCT__entry(
+ __field( void * , group )
+ __field( unsigned int , lvl )
+ ),
+
+ TP_fast_assign(
+ __entry->group = group;
+ __entry->lvl = group->level;
+ ),
+
+ TP_printk("group=%p lvl=%d",
+ __entry->group, __entry->lvl)
+);
+
+#endif /* _TRACE_TIMER_MIGRATION_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/kernel/time/timer_migration.c b/kernel/time/timer_migration.c
index 929b3b94b893..5ac6b4f04c55 100644
--- a/kernel/time/timer_migration.c
+++ b/kernel/time/timer_migration.c
@@ -14,6 +14,9 @@
#include "timer_migration.h"
#include "tick-internal.h"
+#define CREATE_TRACE_POINTS
+#include <trace/events/timer_migration.h>
+
/*
* The timer migration mechanism is built on a hierarchy of groups. The
* lowest level group contains CPUs, the next level groups of CPU groups
@@ -514,6 +517,8 @@ static bool tmigr_active_up(struct tmigr_group *group,
*/
group->groupevt.ignore = true;
+ trace_tmigr_group_set_cpu_active(group, newstate, childmask);
+
return walk_done;
}
@@ -545,6 +550,7 @@ void tmigr_cpu_activate(void)
raw_spin_lock(&tmc->lock);
tmc->idle = false;
+ trace_tmigr_cpu_active(tmc);
__tmigr_cpu_activate(tmc);
raw_spin_unlock(&tmc->lock);
}
@@ -688,6 +694,9 @@ static bool tmigr_update_events(struct tmigr_group *group,
data->nextexp = tmigr_next_groupevt_expires(group);
}
+ trace_tmigr_update_events(child, group, data->childstate,
+ data->groupstate, nextexp);
+
unlock:
raw_spin_unlock(&group->lock);
@@ -721,6 +730,8 @@ static u64 tmigr_new_timer(struct tmigr_cpu *tmc, u64 nextexp)
if (tmc->remote)
return KTIME_MAX;
+ trace_tmigr_cpu_new_timer(tmc);
+
tmc->cpuevt.ignore = false;
data.remote = false;
@@ -754,6 +765,8 @@ static u64 tmigr_handle_remote_cpu(unsigned int cpu, u64 now,
return next;
}
+ trace_tmigr_handle_remote_cpu(tmc);
+
tmc->remote = true;
WRITE_ONCE(tmc->wakeup, KTIME_MAX);
@@ -838,6 +851,7 @@ static bool tmigr_handle_remote_up(struct tmigr_group *group,
childmask = data->childmask;
+ trace_tmigr_handle_remote(group);
again:
/*
* Handle the group only if @childmask is the migrator or if the
@@ -1156,6 +1170,8 @@ static bool tmigr_inactive_up(struct tmigr_group *group,
}
}
+ trace_tmigr_group_set_cpu_inactive(group, newstate, childmask);
+
return walk_done;
}
@@ -1244,6 +1260,7 @@ u64 tmigr_cpu_deactivate(u64 nextexp)
*/
WRITE_ONCE(tmc->wakeup, ret);
+ trace_tmigr_cpu_idle(tmc, nextexp);
raw_spin_unlock(&tmc->lock);
return ret;
}
@@ -1352,6 +1369,7 @@ static struct tmigr_group *tmigr_get_group(unsigned int cpu, int node,
/* Setup successful. Add it to the hierarchy */
list_add(&group->list, &tmigr_level_list[lvl]);
+ trace_tmigr_group_set(group);
return group;
}
@@ -1369,6 +1387,8 @@ static void tmigr_connect_child_parent(struct tmigr_group *child,
raw_spin_unlock(&parent->lock);
raw_spin_unlock_irq(&child->lock);
+ trace_tmigr_connect_child_parent(child);
+
/*
* To prevent inconsistent states, active children need to be active in
* the new parent as well. Inactive children are already marked inactive
@@ -1450,6 +1470,8 @@ static int tmigr_setup_groups(unsigned int cpu, unsigned int node)
raw_spin_unlock_irq(&group->lock);
+ trace_tmigr_connect_cpu_parent(tmc);
+
/* There are no children that need to be connected */
continue;
} else {
@@ -1518,6 +1540,7 @@ static int tmigr_cpu_online(unsigned int cpu)
WRITE_ONCE(tmc->wakeup, KTIME_MAX);
}
raw_spin_lock_irq(&tmc->lock);
+ trace_tmigr_cpu_online(tmc);
__tmigr_cpu_activate(tmc);
tmc->online = true;
raw_spin_unlock_irq(&tmc->lock);
@@ -1537,6 +1560,7 @@ static int tmigr_cpu_offline(unsigned int cpu)
* offline; Therefore nextevt value is set to KTIME_MAX
*/
__tmigr_cpu_deactivate(tmc, KTIME_MAX);
+ trace_tmigr_cpu_offline(tmc);
raw_spin_unlock_irq(&tmc->lock);
return 0;
--
2.39.2
The timer pull model is in place so we can remove the heuristics which try
to guess the best target CPU at enqueue/modification time.
All non pinned timers are queued on the local CPU in the separate storage
and eventually pulled at expiry time to a remote CPU.
Originally-by: Richard Cochran (linutronix GmbH) <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
v6:
- Update TIMER_PINNED flag description.
v5:
- Move WARN_ONCE() in add_timer_on() into a previous patch
- Fold crystallball magic related hunks into this patch
v4: Update comment about TIMER_PINNED flag (heristic is removed)
---
include/linux/timer.h | 14 ++++----------
kernel/time/timer.c | 42 ++++++++++++++++++++----------------------
2 files changed, 24 insertions(+), 32 deletions(-)
diff --git a/include/linux/timer.h b/include/linux/timer.h
index 6f96661480dd..c39ae2ad0035 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -50,16 +50,10 @@ struct timer_list {
* workqueue locking issues. It's not meant for executing random crap
* with interrupts disabled. Abuse is monitored!
*
- * @TIMER_PINNED: A pinned timer will not be affected by any timer
- * placement heuristics (like, NOHZ) and will always expire on the CPU
- * on which the timer was enqueued.
- *
- * Note: Because enqueuing of timers can migrate the timer from one
- * CPU to another, pinned timers are not guaranteed to stay on the
- * initialy selected CPU. They move to the CPU on which the enqueue
- * function is invoked via mod_timer() or add_timer(). If the timer
- * should be placed on a particular CPU, then add_timer_on() has to be
- * used.
+ * @TIMER_PINNED: A pinned timer will always expire on the CPU on which the
+ * timer was enqueued. When a particular CPU is required, add_timer_on()
+ * has to be used. Enqueue via mod_timer() and add_timer() is always done
+ * on the local CPU.
*/
#define TIMER_CPUMASK 0x0003FFFF
#define TIMER_MIGRATING 0x00040000
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 331c59c0789a..feb915642457 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -590,10 +590,13 @@ trigger_dyntick_cpu(struct timer_base *base, struct timer_list *timer)
/*
* We might have to IPI the remote CPU if the base is idle and the
- * timer is not deferrable. If the other CPU is on the way to idle
- * then it can't set base->is_idle as we hold the base lock:
+ * timer is pinned. If it is a non pinned timer, it is only queued
+ * on the remote CPU, when timer was running during queueing. Then
+ * everything is handled by remote CPU anyway. If the other CPU is
+ * on the way to idle then it can't set base->is_idle as we hold
+ * the base lock:
*/
- if (base->is_idle)
+ if (base->is_idle && timer->flags & TIMER_PINNED)
wake_up_nohz_cpu(base->cpu);
}
@@ -941,17 +944,6 @@ static inline struct timer_base *get_timer_base(u32 tflags)
return get_timer_cpu_base(tflags, tflags & TIMER_CPUMASK);
}
-static inline struct timer_base *
-get_target_base(struct timer_base *base, unsigned tflags)
-{
-#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
- if (static_branch_likely(&timers_migration_enabled) &&
- !(tflags & TIMER_PINNED))
- return get_timer_cpu_base(tflags, get_nohz_timer_target());
-#endif
- return get_timer_this_cpu_base(tflags);
-}
-
static inline void __forward_timer_base(struct timer_base *base,
unsigned long basej)
{
@@ -1106,7 +1098,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, unsigned int option
if (!ret && (options & MOD_TIMER_PENDING_ONLY))
goto out_unlock;
- new_base = get_target_base(base, timer->flags);
+ new_base = get_timer_this_cpu_base(timer->flags);
if (base != new_base) {
/*
@@ -2256,8 +2248,14 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
*/
}
- /* We need to mark both bases in sync */
- base_local->is_idle = base_global->is_idle = *idle;
+ /*
+ * base->is_idle information is required to wakeup a idle CPU when
+ * a new timer was enqueued. Only pinned timers could be enqueued
+ * remotely into a idle base. Therefore do maintain only
+ * base_local->is_idle information and ignore base_global->is_idle
+ * information.
+ */
+ base_local->is_idle = *idle;
raw_spin_unlock(&base_global->lock);
raw_spin_unlock(&base_local->lock);
@@ -2273,13 +2271,13 @@ u64 timer_set_idle(unsigned long basej, u64 basem, bool *idle)
void timer_clear_idle(void)
{
/*
- * We do this unlocked. The worst outcome is a remote enqueue sending
- * a pointless IPI, but taking the lock would just make the window for
- * sending the IPI a few instructions smaller for the cost of taking
- * the lock in the exit from idle path.
+ * We do this unlocked. The worst outcome is a remote pinned timer
+ * enqueue sending a pointless IPI, but taking the lock would just
+ * make the window for sending the IPI a few instructions smaller
+ * for the cost of taking the lock in the exit from idle
+ * path. Required for BASE_LOCAL only.
*/
__this_cpu_write(timer_bases[BASE_LOCAL].is_idle, false);
- __this_cpu_write(timer_bases[BASE_GLOBAL].is_idle, false);
/* Activate without holding the timer_base->lock */
tmigr_cpu_activate();
--
2.39.2
There is an already existing function for forwarding the timer
base. Forwarding the timer base is implemented directly in
get_next_timer_interrupt() as well.
To avoid code duplication, replace implementation for forwarding timer base
with the direct call to the already existing function.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
kernel/time/timer.c | 10 ++--------
1 file changed, 2 insertions(+), 8 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 2f6afd1da891..dc58c479d35a 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1939,15 +1939,9 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
/*
* We have a fresh next event. Check whether we can forward the
- * base. We can only do that when @basej is past base->clk
- * otherwise we might rewind base->clk.
+ * base.
*/
- if (time_after(basej, base->clk)) {
- if (time_after(nextevt, basej))
- base->clk = basej;
- else if (time_after(nextevt, base->clk))
- base->clk = nextevt;
- }
+ __forward_timer_base(base, basej);
if (time_before_eq(nextevt, basej)) {
expires = basem;
--
2.39.2
Timer might be used as pinned timer (using add_timer_on()) and later on as
non pinned timers using add_timer(). When the NOHZ timer pull at expiry
model is in place, TIMER_PINNED flag is required to be used whenever a
timer needs to expire on a dedicated CPU. Flag must no be set, if
expiration on a dedicated CPU is not required.
add_timer_on()'s behavior will be changed during the preparation patches
for the NOHZ timer pull at expiry model to unconditionally set TIMER_PINNED
flag. To be able to reset/set the flag when queueing a timer, two variants
of add_timer() are introduced.
This is a preparatory patch and has no functional change.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
---
New in v6
---
include/linux/timer.h | 2 ++
kernel/time/timer.c | 34 ++++++++++++++++++++++++++++++++++
2 files changed, 36 insertions(+)
diff --git a/include/linux/timer.h b/include/linux/timer.h
index 9162f275819a..6f96661480dd 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -180,6 +180,8 @@ extern int timer_reduce(struct timer_list *timer, unsigned long expires);
#define NEXT_TIMER_MAX_DELTA ((1UL << 30) - 1)
extern void add_timer(struct timer_list *timer);
+extern void add_timer_local(struct timer_list *timer);
+extern void add_timer_global(struct timer_list *timer);
extern int try_to_del_timer_sync(struct timer_list *timer);
extern int timer_delete_sync(struct timer_list *timer);
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 8518f7aa7319..3172b8094097 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1245,6 +1245,40 @@ void add_timer(struct timer_list *timer)
}
EXPORT_SYMBOL(add_timer);
+/**
+ * add_timer_local - Start a timer on the local CPU
+ * @timer: The timer to be started
+ *
+ * Same as add_timer() except that the timer flag TIMER_PINNED is set.
+ *
+ * See add_timer() for further details.
+ */
+void add_timer_local(struct timer_list *timer)
+{
+ if (WARN_ON_ONCE(timer_pending(timer)))
+ return;
+ timer->flags |= TIMER_PINNED;
+ __mod_timer(timer, timer->expires, MOD_TIMER_NOTPENDING);
+}
+EXPORT_SYMBOL(add_timer_local);
+
+/**
+ * add_timer_global - Start a timer without TIMER_PINNED flag set
+ * @timer: The timer to be started
+ *
+ * Same as add_timer() except that the timer flag TIMER_PINNED is unset.
+ *
+ * See add_timer() for further details.
+ */
+void add_timer_global(struct timer_list *timer)
+{
+ if (WARN_ON_ONCE(timer_pending(timer)))
+ return;
+ timer->flags &= ~TIMER_PINNED;
+ __mod_timer(timer, timer->expires, MOD_TIMER_NOTPENDING);
+}
+EXPORT_SYMBOL(add_timer_global);
+
/**
* add_timer_on - Start a timer on a particular CPU
* @timer: The timer to be started
--
2.39.2
The functionality for getting the next timer interrupt in
get_next_timer_interrupt() is split into a separate function
fetch_next_timer_interrupt() to be usable by other call sites.
This is preparatory work for the conversion of the NOHZ timer
placement to a pull at expiry time model. No functional change.
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
---
v6: s/splitted/split
v5: Update commit message
v4: Fix typo in comment
---
kernel/time/timer.c | 51 +++++++++++++++++++++++++++------------------
1 file changed, 31 insertions(+), 20 deletions(-)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 4230fc6fa1ed..c3061b28214e 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1976,12 +1976,12 @@ static unsigned long next_timer_interrupt(struct timer_base *base)
return base->next_expiry;
}
-static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64 basem,
- struct timer_base *base_local,
- struct timer_base *base_global,
- struct timer_events *tevt)
+static unsigned long fetch_next_timer_interrupt(unsigned long basej, u64 basem,
+ struct timer_base *base_local,
+ struct timer_base *base_global,
+ struct timer_events *tevt)
{
- unsigned long nextevt, nextevt_local, nextevt_global;
+ unsigned long nextevt_local, nextevt_global;
bool local_first;
nextevt_local = next_timer_interrupt(base_local);
@@ -2000,21 +2000,6 @@ static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64
else
local_first = base_local->timers_pending;
- nextevt = local_first ? nextevt_local : nextevt_global;
-
- /*
- * If the @nextevt is at max. one tick away, use @nextevt and store
- * it in the local expiry value. The next global event is irrelevant in
- * this case and can be left as KTIME_MAX.
- */
- if (time_before_eq(nextevt, basej + 1)) {
- /* If we missed a tick already, force 0 delta */
- if (time_before(nextevt, basej))
- nextevt = basej;
- tevt->local = basem + (u64)(nextevt - basej) * TICK_NSEC;
- return tevt->local;
- }
-
/*
* Update tevt->* values:
*
@@ -2028,6 +2013,32 @@ static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64
if (base_local->timers_pending)
tevt->local = basem + (u64)(nextevt_local - basej) * TICK_NSEC;
+ return local_first ? nextevt_local : nextevt_global;
+}
+
+static inline unsigned long __get_next_timer_interrupt(unsigned long basej, u64 basem,
+ struct timer_base *base_local,
+ struct timer_base *base_global,
+ struct timer_events *tevt)
+{
+ unsigned long nextevt;
+
+ nextevt = fetch_next_timer_interrupt(basej, basem,
+ base_local, base_global, tevt);
+
+ /*
+ * If the nextevt is at max. one tick away, use the @nextevt and store
+ * it in the local expiry value. Reset of next global event to KTIME_MAX
+ * to make sure it is ignored.
+ */
+ if (time_before_eq(nextevt, basej + 1)) {
+ /* If we missed a tick already, force 0 delta */
+ if (time_before(nextevt, basej))
+ nextevt = basej;
+ tevt->local = basem + (u64)(nextevt - basej) * TICK_NSEC;
+ tevt->global = KTIME_MAX;
+ }
+
return nextevt;
}
--
2.39.2
Hello Anna-Maria,
On 10/4/2023 6:04 PM, Anna-Maria Behnsen wrote:
> [..snip..]
>
> Ping Pong Oberservation
> ^^^^^^^^^^^^^^^^^^^^^^^
>
> During testing on a mostly idle machine a ping pong game could be observed:
> a process_timeout timer is expired remotely on a non idle CPU. Then the CPU
> where the schedule_timeout() was executed to enqueue the timer comes out of
> idle and restarts the timer using schedule_timeout() and goes back to idle
> again. This is due to the fair scheduler which tries to keep the task on
> the CPU which it previously executed on.
Regarding above, are you referring to "wake_up_process(timeout->task)" in
"process_timeout()" ends up waking the task on an idle CPU instead of the
CPU where process_timeout() ran?
In which case, have you tried using the "WF_CURRENT_CPU" flag for the
wakeup? (landed upstream in v6.6-rc1) It is only used by wait queues in
kernel/sched/wait.c currently but perhaps we can have a
"wake_up_process_on_current_cpu()" that process_timeout() can call.
Something along the lines of:
int wake_up_process_on_current_cpu(struct task_struct *p)
{
return try_to_wake_up(p, TASK_NORMAL, WF_CURRENT_CPU);
}
EXPORT_SYMBOL(wake_up_process_on_current_cpu);
Thoughts?
>
> [..snip..]
>
--
Thanks and Regards,
Prateek
On Wed, Oct 04, 2023 at 02:34:36PM +0200, Anna-Maria Behnsen wrote:
> There is an already existing function for forwarding the timer
> base. Forwarding the timer base is implemented directly in
> get_next_timer_interrupt() as well.
>
> To avoid code duplication, replace implementation for forwarding timer base
> with the direct call to the already existing function.
>
> Signed-off-by: Anna-Maria Behnsen <[email protected]>
Good catch!
Reviewed-by: Frederic Weisbecker <[email protected]>
On Wed, Oct 04 2023 at 14:34, Anna-Maria Behnsen wrote:
>
> - if (time_before_eq(nextevt, basej)) {
> - expires = basem;
> - base->is_idle = false;
> - } else {
> - if (base->timers_pending)
> - expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
> - /*
> - * If we expect to sleep more than a tick, mark the base idle.
> - * Also the tick is stopped so any added timer must forward
> - * the base clk itself to keep granularity small. This idle
> - * logic is only maintained for the BASE_STD base, deferrable
> - * timers may still see large granularity skew (by design).
> - */
> - if ((expires - basem) > TICK_NSEC)
> - base->is_idle = true;
> + /*
> + * Base is idle if the next event is more than a tick away. Also
> + * the tick is stopped so any added timer must forward the base clk
> + * itself to keep granularity small. This idle logic is only
> + * maintained for the BASE_STD base, deferrable timers may still
> + * see large granularity skew (by design).
> + */
> + base->is_idle = time_after(nextevt, basej + 1);
This is wrongly ordered. base->is_idle must be updated _after_
evaluating base->timers_pending because the below can change nextevt,
no?
> + if (base->timers_pending) {
> + /* If we missed a tick already, force 0 delta */
> + if (time_before(nextevt, basej))
> + nextevt = basej;
> + expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
Thanks,
tglx
On Tue, Oct 10, 2023 at 12:15:09AM +0200, Thomas Gleixner wrote:
> On Wed, Oct 04 2023 at 14:34, Anna-Maria Behnsen wrote:
> >
> > - if (time_before_eq(nextevt, basej)) {
> > - expires = basem;
> > - base->is_idle = false;
> > - } else {
> > - if (base->timers_pending)
> > - expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
> > - /*
> > - * If we expect to sleep more than a tick, mark the base idle.
> > - * Also the tick is stopped so any added timer must forward
> > - * the base clk itself to keep granularity small. This idle
> > - * logic is only maintained for the BASE_STD base, deferrable
> > - * timers may still see large granularity skew (by design).
> > - */
> > - if ((expires - basem) > TICK_NSEC)
> > - base->is_idle = true;
> > + /*
> > + * Base is idle if the next event is more than a tick away. Also
> > + * the tick is stopped so any added timer must forward the base clk
> > + * itself to keep granularity small. This idle logic is only
> > + * maintained for the BASE_STD base, deferrable timers may still
> > + * see large granularity skew (by design).
> > + */
> > + base->is_idle = time_after(nextevt, basej + 1);
>
> This is wrongly ordered. base->is_idle must be updated _after_
> evaluating base->timers_pending because the below can change nextevt,
> no?
>
> > + if (base->timers_pending) {
> > + /* If we missed a tick already, force 0 delta */
> > + if (time_before(nextevt, basej))
> > + nextevt = basej;
> > + expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
I suspect it doesn't matter in pratice: base->is_idle will remain false
if it's before/equal jiffies.
Still it hurts the eyes so I agree the re-ordering should happen here and
this will even simplify a bit the next patch.
Thanks.
> Thanks,
>
> tglx
On Tue, Oct 10 2023 at 13:19, Frederic Weisbecker wrote:
> On Tue, Oct 10, 2023 at 12:15:09AM +0200, Thomas Gleixner wrote:
>> > + base->is_idle = time_after(nextevt, basej + 1);
>>
>> This is wrongly ordered. base->is_idle must be updated _after_
>> evaluating base->timers_pending because the below can change nextevt,
>> no?
>>
>> > + if (base->timers_pending) {
>> > + /* If we missed a tick already, force 0 delta */
>> > + if (time_before(nextevt, basej))
>> > + nextevt = basej;
>> > + expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
>
> I suspect it doesn't matter in pratice: base->is_idle will remain false
> if it's before/equal jiffies.
>
> Still it hurts the eyes so I agree the re-ordering should happen here and
> this will even simplify a bit the next patch.
Right. Anna-Maria just pointed that out to me before, but we are all in
violent agreement that it sucks :)
Hi Maria,
On Wed, 2023-10-04 at 14:34 +0200, Anna-Maria Behnsen wrote:
> Hi,
>
>
[...]
>
> The proper solution to this problem is to always queue the timers on
> the
> local CPU and allow the non pinned timers to be pulled onto a busy
> CPU at
> expiry time.
Thanks for these patches. I am looking for saving power during video
playback with our low power daemon. I use cgroup v2 isolation to keep
some CPUs idle (CPU 0-11) and video is played on a single module (CPU
12-15).
I have some kernelshark pictures at below link. The traces are
collected with sched, timer and irq. With 6.6-rc5, you can see some
timers still expires on CPUs which I want to keep idle. With timer
patches added, they are mostly pulled to busy CPU.
https://imgur.com/a/8nF5OoP
I can share the .dat files, but they are too big to attach here.
Thanks,
Srinivas
>
> Therefore split the timer storage into local pinned and global
> timers:
> Local pinned timers are always expired on the CPU on which they have
> been
> queued. Global timers can be expired on any CPU.
>
> As long as a CPU is busy it expires both local and global timers.
> When a
> CPU goes idle it arms for the first expiring local timer. If the
> first
> expiring pinned (local) timer is before the first expiring movable
> timer,
> then no action is required because the CPU will wake up before the
> first
> movable timer expires. If the first expiring movable timer is before
> the
> first expiring pinned (local) timer, then this timer is queued into a
> idle
> timerqueue and eventually expired by some other active CPU.
>
> To avoid global locking the timerqueues are implemented as a
> hierarchy. The
> lowest level of the hierarchy holds the CPUs. The CPUs are associated
> to
> groups of 8, which are separated per node. If more than one CPU group
> exist, then a second level in the hierarchy collects the groups.
> Depending
> on the size of the system more than 2 levels are required. Each group
> has a
> "migrator" which checks the timerqueue during the tick for remote
> timers to
> be expired.
>
> If the last CPU in a group goes idle it reports the first expiring
> event in
> the group up to the next group(s) in the hierarchy. If the last CPU
> goes
> idle it arms its timer for the first system wide expiring timer to
> ensure
> that no timer event is missed.
>
>
> Testing
> ~~~~~~~
>
> Enqueue
> ^^^^^^^
>
> The impact of wasting cycles during enqueue by using the heuristic in
> contrast to always queuing the timer on the local CPU was measured
> with a
> micro benchmark. Therefore a timer is enqueued and dequeued in a loop
> with
> 1000 repetitions on a isolated CPU. The time the loop takes is
> measured. A
> quarter of the remaining CPUs was kept busy. This measurement was
> repeated
> several times. With the patch queue the average duration was reduced
> by
> approximately 25%.
>
> 145ns plain v6
> 109ns v6 with patch queue
>
>
> Furthermore the impact of residence in deep idle states of an idle
> system
> was investigated. The patch queue doesn't downgrade this behavior.
>
> dbench test
> ^^^^^^^^^^^
>
> A dbench test starting X pairs of client servers are used to create
> load on
> the system. The measurable value is the throughput. The tests were
> executed
> on a zen3 machine. The base is the tip tree branch timers/core which
> is
> based on a v6.6-rc1.
>
> governor menu
>
> X pairs timers/core pull-model impact
> ----------------------------------------------
> 1 353.19 (0.19) 353.45 (0.30) 0.07%
> 2 700.10 (0.96) 687.00 (0.20) -1.87%
> 4 1329.37 (0.63) 1282.91 (0.64) -3.49%
> 8 2561.16 (1.28) 2493.56 (1.76) -2.64%
> 16 4959.96 (0.80) 4914.59 (0.64) -0.91%
> 32 9741.92 (3.44) 8979.83 (1.13) -7.82%
> 64 16535.40 (2.84) 16388.47 (4.02) -0.89%
> 128 22136.83 (2.42) 23174.50 (1.43) 4.69%
> 256 39256.77 (4.48) 38994.00 (0.39) -0.67%
> 512 36799.03 (1.83) 38091.10 (0.63) 3.51%
> 1024 32903.03 (0.86) 35370.70 (0.89) 7.50%
>
>
> governor teo
>
> X pairs timers/core pull-model impact
> ----------------------------------------------
> 1 350.83 (1.27) 352.45 (0.96) 0.46%
> 2 699.52 (0.85) 690.10 (0.54) -1.35%
> 4 1339.53 (1.99) 1294.71 (2.71) -3.35%
> 8 2574.10 (0.76) 2495.46 (1.97) -3.06%
> 16 4898.50 (1.74) 4783.06 (1.64) -2.36%
> 32 9115.50 (4.63) 9037.83 (1.58) -0.85%
> 64 16663.90 (3.80) 16042.00 (1.72) -3.73%
> 128 25044.93 (1.11) 23250.03 (1.08) -7.17%
> 256 38059.53 (1.70) 39658.57 (2.98) 4.20%
> 512 36369.30 (0.39) 38890.13 (0.36) 6.93%
> 1024 33956.83 (1.14) 35514.83 (0.29) 4.59%
>
>
>
> Ping Pong Oberservation
> ^^^^^^^^^^^^^^^^^^^^^^^
>
> During testing on a mostly idle machine a ping pong game could be
> observed:
> a process_timeout timer is expired remotely on a non idle CPU. Then
> the CPU
> where the schedule_timeout() was executed to enqueue the timer comes
> out of
> idle and restarts the timer using schedule_timeout() and goes back to
> idle
> again. This is due to the fair scheduler which tries to keep the task
> on
> the CPU which it previously executed on.
>
>
>
>
> Possible Next Steps
> ~~~~~~~~~~~~~~~~~~~
>
> Simple deferrable timers are no longer required as they can be
> converted to
> global timers. If a CPU goes idle, a formerly deferrable timer will
> not
> prevent the CPU to sleep as long as possible. Only the last migrator
> CPU
> has to take care of them. Deferrable timers with timer pinned flags
> needs
> to be expired on the specified CPU but must not prevent CPU from
> going
> idle. They require their own timer base which is never taken into
> account
> when calculating the next expiry time. This conversation and required
> cleanup will be done in a follow up series.
>
>
> v7..v8:
> https://lore.kernel.org/r/[email protected]
> - Address review feedback
> - Move marking timer base idle into tick_nohz_stop_tick()
> - Look ahead function to determine possible sleep lenght
>
>
> v6..v7:
> - Address review feedback of Frederic and bigeasy
> - Change lock, unlock fetch next timer interrupt logic after remote
> expiry
> - Move timer_expire_remote() into tick-internal.h
> - Add documentation section about "Required event and timerqueue
> update
> after remote expiry"
> - Fix fallout of kernel test robot
>
>
> v5..v6:
>
> - Address review of Frederic Weisbecker and Peter Zijlstra
> (spelling,
> locking, race in tmigr_handle_remote_cpu())
>
> - unconditionally set TIMER_PINNED flag in add_timer_on();
> introduce
> add_timer() variants which set/unset TIMER_PINNED flag; drop
> fixing
> add_timer_on() call sites, as TIMER_PINNED flag is set
> implicitly;
> Fixing workqueue to use add_timer_global() instead of simply
> add_timer() for unbound work.
>
> - Drop support for siblings to end up in the same level 0 group
> (could be
> added again in a better way as an improvement later on)
>
> - Do not send IPI for new first deferrable timers
>
> v4..v5:
> - address review feedback of Frederic Weisbecker
> - fix issue with group timer update after remote expiry
>
> v3..v4:
> - address review feedback of Frederic Weisbecker
> - address kernel test robot fallout
> - Move patch 16 "add_timer_on(): Make sure callers have
> TIMER_PINNED
> flag" at the begin of the queue to prevent timers to end up in
> global
> timer base when they were queued using add_timer_on()
> - Fix some comments and typos
>
> v2..v3:
> https://lore.kernel.org/r/[email protected]/
> - Minimize usage of locks by storing data using atomic_cmpxchg()
> for
> migrator information and information about active cpus.
>
>
> Thanks,
>
> Anna-Maria
>
>
>
> Anna-Maria Behnsen (22):
> tick/sched: Cleanup confusing variables
> tick-sched: Warn when next tick seems to be in the past
> timer: Do not IPI for deferrable timers
> timer: Move store of next event into __next_timer_interrupt()
> timers: Clarify check in forward_timer_base()
> timers: Split out forward timer base functionality
> timers: Use already existing function for forwarding timer base
> timer: Split out get next timer functionality
> timers: Move marking timer bases idle into tick_nohz_stop_tick()
> timers: Introduce add_timer() variants which modify timer flags
> workqueue: Use global variant for add_timer()
> timer: add_timer_on(): Make sure TIMER_PINNED flag is set
> timers: Ease code in run_local_timers()
> timer: Split next timer interrupt logic
> timer: Keep the pinned timers separate from the others
> timer: Retrieve next expiry of pinned/non-pinned timers separately
> timer: Split out "get next timer interrupt" functionality
> timer: Add get next timer interrupt functionality for remote CPUs
> timer: Check if timers base is handled already
> timer: Implement the hierarchical pull model
> timer_migration: Add tracepoints
> timer: Always queue timers on the local CPU
>
> Richard Cochran (linutronix GmbH) (2):
> timer: Restructure internal locking
> tick/sched: Split out jiffies update helper function
>
> Thomas Gleixner (1):
> timer: Rework idle logic
>
> include/linux/cpuhotplug.h | 1 +
> include/linux/timer.h | 16 +-
> include/trace/events/timer_migration.h | 283 ++++
> kernel/time/Makefile | 3 +
> kernel/time/tick-internal.h | 13 +
> kernel/time/tick-sched.c | 69 +-
> kernel/time/timer.c | 514 ++++++--
> kernel/time/timer_migration.c | 1636
> ++++++++++++++++++++++++
> kernel/time/timer_migration.h | 144 +++
> kernel/workqueue.c | 2 +-
> 10 files changed, 2549 insertions(+), 132 deletions(-)
> create mode 100644 include/trace/events/timer_migration.h
> create mode 100644 kernel/time/timer_migration.c
> create mode 100644 kernel/time/timer_migration.h
>
Hello Anna-Maria,
Happy to report I don't see any regression with this version of series.
I'll leave the detailed report below.
On 10/4/2023 6:04 PM, Anna-Maria Behnsen wrote:
> [..snip..]
>
> dbench test
> ^^^^^^^^^^^
>
> A dbench test starting X pairs of client servers are used to create load on
> the system. The measurable value is the throughput. The tests were executed
> on a zen3 machine. The base is the tip tree branch timers/core which is
> based on a v6.6-rc1.
>
> governor menu
>
> X pairs timers/core pull-model impact
> ----------------------------------------------
> 1 353.19 (0.19) 353.45 (0.30) 0.07%
> 2 700.10 (0.96) 687.00 (0.20) -1.87%
> 4 1329.37 (0.63) 1282.91 (0.64) -3.49%
> 8 2561.16 (1.28) 2493.56 (1.76) -2.64%
> 16 4959.96 (0.80) 4914.59 (0.64) -0.91%
> 32 9741.92 (3.44) 8979.83 (1.13) -7.82%
> 64 16535.40 (2.84) 16388.47 (4.02) -0.89%
> 128 22136.83 (2.42) 23174.50 (1.43) 4.69%
> 256 39256.77 (4.48) 38994.00 (0.39) -0.67%
> 512 36799.03 (1.83) 38091.10 (0.63) 3.51%
> 1024 32903.03 (0.86) 35370.70 (0.89) 7.50%
>
>
> governor teo
>
> X pairs timers/core pull-model impact
> ----------------------------------------------
> 1 350.83 (1.27) 352.45 (0.96) 0.46%
> 2 699.52 (0.85) 690.10 (0.54) -1.35%
> 4 1339.53 (1.99) 1294.71 (2.71) -3.35%
> 8 2574.10 (0.76) 2495.46 (1.97) -3.06%
> 16 4898.50 (1.74) 4783.06 (1.64) -2.36%
> 32 9115.50 (4.63) 9037.83 (1.58) -0.85%
> 64 16663.90 (3.80) 16042.00 (1.72) -3.73%
> 128 25044.93 (1.11) 23250.03 (1.08) -7.17%
> 256 38059.53 (1.70) 39658.57 (2.98) 4.20%
> 512 36369.30 (0.39) 38890.13 (0.36) 6.93%
> 1024 33956.83 (1.14) 35514.83 (0.29) 4.59%
o Machine details
- 3rd Generation EPYC System
- 2 sockets each with 64C/128T
- NPS1 (Each socket is a NUMA node)
- C2 Disabled (POLL and C1(MWAIT) remained enabled)
o Kernel Details
- tip: tip:sched/core at commit 238437d88cea ("intel_idle: Add ibrs_off
module parameter to force-disable IBRS") + min_deadline fix
commit 8dafa9d0eb1a ("sched/eevdf: Fix min_deadline heap
integrity") from tip:sched/urgent
- timer-pull: tip + this series as is
o Benchmark Results
==================================================================
Test : hackbench
Units : Normalized time in seconds
Interpretation: Lower is better
Statistic : AMean
==================================================================
Case: tip[pct imp](CV) timer-pull[pct imp](CV)
1-groups 1.00 [ -0.00]( 2.11) 0.99 [ 1.44]( 3.34)
2-groups 1.00 [ -0.00]( 1.31) 1.01 [ -0.93]( 1.57)
4-groups 1.00 [ -0.00]( 1.04) 1.00 [ 0.44]( 1.11)
8-groups 1.00 [ -0.00]( 1.34) 0.99 [ 1.29]( 1.34)
16-groups 1.00 [ -0.00]( 2.45) 1.00 [ -0.40]( 2.78)
==================================================================
Test : tbench
Units : Normalized throughput
Interpretation: Higher is better
Statistic : AMean
==================================================================
Clients: tip[pct imp](CV) timer-pull[pct imp](CV)
1 1.00 [ 0.00]( 0.46) 1.01 [ 0.52]( 0.66)
2 1.00 [ 0.00]( 0.64) 0.99 [ -0.60]( 0.88)
4 1.00 [ 0.00]( 0.59) 0.99 [ -0.92]( 1.82)
8 1.00 [ 0.00]( 0.34) 1.00 [ -0.06]( 0.33)
16 1.00 [ 0.00]( 0.72) 0.99 [ -1.25]( 1.52)
32 1.00 [ 0.00]( 0.65) 0.98 [ -1.59]( 1.29)
64 1.00 [ 0.00]( 0.59) 0.99 [ -0.84]( 3.87)
128 1.00 [ 0.00]( 1.19) 1.00 [ 0.11]( 0.33)
256 1.00 [ 0.00]( 0.16) 1.01 [ 0.61]( 0.52)
512 1.00 [ 0.00]( 0.20) 1.01 [ 0.80]( 0.29)
1024 1.00 [ 0.00]( 0.06) 1.01 [ 1.06]( 0.59)
==================================================================
Test : stream-10
Units : Normalized Bandwidth, MB/s
Interpretation: Higher is better
Statistic : HMean
==================================================================
Test: tip[pct imp](CV) timer-pull[pct imp](CV)
Copy 1.00 [ 0.00]( 6.04) 1.04 [ 4.31]( 3.71)
Scale 1.00 [ 0.00]( 5.44) 1.01 [ 0.57]( 5.63)
Add 1.00 [ 0.00]( 5.44) 1.01 [ 0.99]( 5.46)
Triad 1.00 [ 0.00]( 7.82) 1.04 [ 4.14]( 5.68)
==================================================================
Test : stream-100
Units : Normalized Bandwidth, MB/s
Interpretation: Higher is better
Statistic : HMean
==================================================================
Test: tip[pct imp](CV) timer-pull[pct imp](CV)
Copy 1.00 [ 0.00]( 1.14) 1.00 [ 0.29]( 0.49)
Scale 1.00 [ 0.00]( 4.60) 1.03 [ 2.87]( 0.62)
Add 1.00 [ 0.00]( 4.91) 1.01 [ 1.36]( 1.34)
Triad 1.00 [ 0.00]( 0.60) 0.98 [ -1.50]( 4.24)
==================================================================
Test : netperf
Units : Normalized Througput
Interpretation: Higher is better
Statistic : AMean
==================================================================
Clients: tip[pct imp](CV) timer-pull[pct imp](CV)
1-clients 1.00 [ 0.00]( 0.61) 1.01 [ 1.25]( 0.48)
2-clients 1.00 [ 0.00]( 0.44) 1.00 [ 0.34]( 0.65)
4-clients 1.00 [ 0.00]( 0.75) 1.01 [ 0.98]( 1.26)
8-clients 1.00 [ 0.00]( 0.65) 1.01 [ 0.82]( 0.73)
16-clients 1.00 [ 0.00]( 0.49) 1.00 [ 0.37]( 0.99)
32-clients 1.00 [ 0.00]( 0.57) 0.98 [ -2.05]( 3.44)
64-clients 1.00 [ 0.00]( 1.67) 1.00 [ 0.00]( 1.74)
128-clients 1.00 [ 0.00]( 1.11) 1.01 [ 0.69]( 1.11)
256-clients 1.00 [ 0.00]( 2.64) 1.00 [ 0.00]( 3.79)
512-clients 1.00 [ 0.00](52.49) 1.00 [ 0.26](54.13)
==================================================================
Test : schbench
Units : Normalized 99th percentile latency in us
Interpretation: Lower is better
Statistic : Median
==================================================================
#workers: tip[pct imp](CV) timer-pull[pct imp](CV)
1 1.00 [ -0.00]( 8.41) 0.59 [ 40.54](40.25)
2 1.00 [ -0.00]( 5.29) 0.93 [ 7.50]( 9.01)
4 1.00 [ -0.00]( 1.32) 0.91 [ 9.09](12.33)
8 1.00 [ -0.00]( 9.52) 1.00 [ -0.00](15.02)
16 1.00 [ -0.00]( 1.61) 1.03 [ -3.23]( 2.37)
32 1.00 [ -0.00]( 7.27) 0.92 [ 7.69]( 1.59)
64 1.00 [ -0.00]( 6.96) 1.12 [-11.56]( 1.20)
128 1.00 [ -0.00]( 3.41) 1.06 [ -6.49]( 3.73)
256 1.00 [ -0.00](32.95) 1.02 [ -2.48](28.66)
512 1.00 [ -0.00]( 3.20) 0.99 [ 0.71]( 3.22)
==================================================================
Test : ycsb-cassandra
Units : Normalized throughput
Interpretation: Higher is better
Statistic : Mean
==================================================================
metric tip timer-pull (%diff)
throughput 1.00 1.01 (%diff: 0.75%)
==================================================================
Test : ycsb-mondodb
Units : Normalized throughput
Interpretation: Higher is better
Statistic : Mean
==================================================================
metric tip timer-pull (%diff)
throughput 1.00 1.00 (%diff: -0.49%)
==================================================================
Test : DeathStarBench
Units : Normalized throughput
Interpretation: Higher is better
Statistic : Mean
==================================================================
Pinning scaling tip timer-pull (%diff)
1CCD 1 1.00 1.01 (%diff: 0.75%)
2CCD 2 1.00 1.03 (%diff: 2.72%)
4CCD 4 1.00 1.00 (%diff: -0.28%)
8CCD 8 1.00 1.00 (%diff: 0.20%)
--
Thank you for debugging and helping fix the tbench regression.
If the series does not change drastically, feel free to add:
Tested-by: K Prateek Nayak <[email protected]>
>
>
>
> Ping Pong Oberservation
> ^^^^^^^^^^^^^^^^^^^^^^^
>
> During testing on a mostly idle machine a ping pong game could be observed:
> a process_timeout timer is expired remotely on a non idle CPU. Then the CPU
> where the schedule_timeout() was executed to enqueue the timer comes out of
> idle and restarts the timer using schedule_timeout() and goes back to idle
> again. This is due to the fair scheduler which tries to keep the task on
> the CPU which it previously executed on.
>
>
>
>
> Possible Next Steps
> ~~~~~~~~~~~~~~~~~~~
>
> Simple deferrable timers are no longer required as they can be converted to
> global timers. If a CPU goes idle, a formerly deferrable timer will not
> prevent the CPU to sleep as long as possible. Only the last migrator CPU
> has to take care of them. Deferrable timers with timer pinned flags needs
> to be expired on the specified CPU but must not prevent CPU from going
> idle. They require their own timer base which is never taken into account
> when calculating the next expiry time. This conversation and required
> cleanup will be done in a follow up series.
>
I'll keep an eye out for future versions for testing.
>
> [..snip..]
>
--
Thanks and Regards,
Prateek
Le Wed, Oct 04, 2023 at 02:34:39PM +0200, Anna-Maria Behnsen a ?crit :
> static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
> {
> struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
> + unsigned long basejiff = ts->last_jiffies;
> u64 basemono = ts->timer_expires_base;
> - u64 expires = ts->timer_expires;
> + bool timer_idle = ts->tick_stopped;
> + u64 expires;
>
> /* Make sure we won't be trying to stop it twice in a row. */
> ts->timer_expires_base = 0;
>
> + /*
> + * Now the tick should be stopped definitely - so timer base needs to be
> + * marked idle as well to not miss a newly queued timer.
> + */
> + expires = timer_set_idle(basejiff, basemono, &timer_idle);
> + if (!timer_idle) {
> + /*
> + * Do not clear tick_stopped here when it was already set - it will
> + * be retained on next idle iteration when tick expired earlier
> + * than expected.
> + */
> + expires = basemono + TICK_NSEC;
> +
> + /* Undo the effect of timer_set_idle() */
> + timer_clear_idle();
Looks like you don't even need to clear ->is_idle on failure. timer_set_idle()
does it for you.
> + } else if (expires < ts->timer_expires) {
> + ts->timer_expires = expires;
> + } else {
> + expires = ts->timer_expires;
Is it because timer_set_idle() doesn't recalculate the next hrtimer (as opposed
to get_next_timer_interrupt())? And since tick_nohz_next_event() did, the fact
that ts->timer_expires has a lower value may mean there is an hrtimer to take
into account and so you rather use the old calculation?
If so please add a comment explaining that because it's not that obvious. It's
worth noting also the side effect that the nearest timer may have been cancelled
in-between and we might reprogram too-early but the event should be rare enough
that we don't care.
Another reason also is that cpuidle may have programmed a shallow C-state
because it saw an early next expiration estimation. And if the related timer is
cancelled in-between and we didn't keep the old expiration estimation, we would
otherwise stop the tick for a long time with a shallow C-state.
> @@ -926,7 +944,7 @@ static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
> * first call we save the current tick time, so we can restart
> * the scheduler tick in nohz_restart_sched_tick.
> */
> - if (!ts->tick_stopped) {
> + if (!ts->tick_stopped && timer_idle) {
In fact, if (!ts->tick_stopped && !timer_idle) then you
should return now and avoid the reprogramming.
> @@ -1950,6 +1950,40 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
> if (cpu_is_offline(smp_processor_id()))
> return expires;
>
> + raw_spin_lock(&base->lock);
> + nextevt = __get_next_timer_interrupt(basej, base);
> + raw_spin_unlock(&base->lock);
It's unfortunate we have to lock here, which means we lock twice
on the idle path. But I can't think of a better way and I guess
the follow-up patches rely on that.
Thanks.
Hi Anna-Maria
On 10/4/23 13:34, Anna-Maria Behnsen wrote:
> Hi,
>
[snip]
>
>
> Testing
> ~~~~~~~
>
> Enqueue
> ^^^^^^^
>
> The impact of wasting cycles during enqueue by using the heuristic in
> contrast to always queuing the timer on the local CPU was measured with a
> micro benchmark. Therefore a timer is enqueued and dequeued in a loop with
> 1000 repetitions on a isolated CPU. The time the loop takes is measured. A
> quarter of the remaining CPUs was kept busy. This measurement was repeated
> several times. With the patch queue the average duration was reduced by
> approximately 25%.
>
> 145ns plain v6
> 109ns v6 with patch queue
>
>
> Furthermore the impact of residence in deep idle states of an idle system
> was investigated. The patch queue doesn't downgrade this behavior.
>
> dbench test
> ^^^^^^^^^^^
>
> A dbench test starting X pairs of client servers are used to create load on
> the system. The measurable value is the throughput. The tests were executed
> on a zen3 machine. The base is the tip tree branch timers/core which is
> based on a v6.6-rc1.
>
> governor menu
>
> X pairs timers/core pull-model impact
> ----------------------------------------------
> 1 353.19 (0.19) 353.45 (0.30) 0.07%
> 2 700.10 (0.96) 687.00 (0.20) -1.87%
> 4 1329.37 (0.63) 1282.91 (0.64) -3.49%
> 8 2561.16 (1.28) 2493.56 (1.76) -2.64%
> 16 4959.96 (0.80) 4914.59 (0.64) -0.91%
> 32 9741.92 (3.44) 8979.83 (1.13) -7.82%
> 64 16535.40 (2.84) 16388.47 (4.02) -0.89%
> 128 22136.83 (2.42) 23174.50 (1.43) 4.69%
> 256 39256.77 (4.48) 38994.00 (0.39) -0.67%
> 512 36799.03 (1.83) 38091.10 (0.63) 3.51%
> 1024 32903.03 (0.86) 35370.70 (0.89) 7.50%
>
>
> governor teo
>
> X pairs timers/core pull-model impact
> ----------------------------------------------
> 1 350.83 (1.27) 352.45 (0.96) 0.46%
> 2 699.52 (0.85) 690.10 (0.54) -1.35%
> 4 1339.53 (1.99) 1294.71 (2.71) -3.35%
> 8 2574.10 (0.76) 2495.46 (1.97) -3.06%
> 16 4898.50 (1.74) 4783.06 (1.64) -2.36%
> 32 9115.50 (4.63) 9037.83 (1.58) -0.85%
> 64 16663.90 (3.80) 16042.00 (1.72) -3.73%
> 128 25044.93 (1.11) 23250.03 (1.08) -7.17%
> 256 38059.53 (1.70) 39658.57 (2.98) 4.20%
> 512 36369.30 (0.39) 38890.13 (0.36) 6.93%
> 1024 33956.83 (1.14) 35514.83 (0.29) 4.59%
>
>
>
> Ping Pong Oberservation
> ^^^^^^^^^^^^^^^^^^^^^^^
>
> During testing on a mostly idle machine a ping pong game could be observed:
> a process_timeout timer is expired remotely on a non idle CPU. Then the CPU
> where the schedule_timeout() was executed to enqueue the timer comes out of
> idle and restarts the timer using schedule_timeout() and goes back to idle
> again. This is due to the fair scheduler which tries to keep the task on
> the CPU which it previously executed on.
>
>
I have tested this on my 2 Arm boards with mainline kernel
and almost-mainline. On both platforms it looks stable.
The results w/ your patchset looks better.
1. rockpi4b - mainline kernel (but no UI)
Limiting the cpumask for only 4 Little CPUs and setting
performance governor for cpufreq and menu for idle.
1.1. perf bench sched pipe
w/o patchset vs. w/ patchset
avg [ops/sec]:
(more is better)
23012.33 vs. 23154.33 (+0.6%)
avg [usecs/op]:
(less is better)
43.453 vs. 43.187 (-0.6%)
1.2. perf bench sched messaging
(less is better)
w/o patchset vs. w/ patchset
avg total time [s]:
2.7855 vs. 2.7005 (-3.1%)
2. pixel6 (kernel v5.18 with backported patchset)
2.1 Speedometer 2.0 (JS test running in Chrome browser)
w/o patchset vs. w/ patchset
149 vs. 146 (-2%)
2.2 Geekbench 5
(more is better)
Single core
w/o patchset vs. w/ patchset
1025 vs. 1017 (-0.7%)
Multi core
w/o patchset vs. w/ patchset
2756 vs. 2813 (+2%)
The performance looks good. Only one test 'Speedometer'
has some interesting lower score.
Fill free to add:
Tested-by: Lukasz Luba <[email protected]>
Regards,
Lukasz
Frederic Weisbecker <[email protected]> writes:
> Le Wed, Oct 04, 2023 at 02:34:39PM +0200, Anna-Maria Behnsen a écrit :
>> static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
>> {
>> struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
>> + unsigned long basejiff = ts->last_jiffies;
>> u64 basemono = ts->timer_expires_base;
>> - u64 expires = ts->timer_expires;
>> + bool timer_idle = ts->tick_stopped;
>> + u64 expires;
>>
>> /* Make sure we won't be trying to stop it twice in a row. */
>> ts->timer_expires_base = 0;
>>
>> + /*
>> + * Now the tick should be stopped definitely - so timer base needs to be
>> + * marked idle as well to not miss a newly queued timer.
>> + */
>> + expires = timer_set_idle(basejiff, basemono, &timer_idle);
>> + if (!timer_idle) {
>> + /*
>> + * Do not clear tick_stopped here when it was already set - it will
>> + * be retained on next idle iteration when tick expired earlier
>> + * than expected.
>> + */
>> + expires = basemono + TICK_NSEC;
>> +
>> + /* Undo the effect of timer_set_idle() */
>> + timer_clear_idle();
>
> Looks like you don't even need to clear ->is_idle on failure. timer_set_idle()
> does it for you.
You are right. I tried several approaches and then forgot to remove it
here.
>> + } else if (expires < ts->timer_expires) {
>> + ts->timer_expires = expires;
>> + } else {
>> + expires = ts->timer_expires;
>
> Is it because timer_set_idle() doesn't recalculate the next hrtimer (as opposed
> to get_next_timer_interrupt())? And since tick_nohz_next_event() did, the fact
> that ts->timer_expires has a lower value may mean there is an hrtimer to take
> into account and so you rather use the old calculation?
Yes and because power things rely on it.
> If so please add a comment explaining that because it's not that obvious. It's
> worth noting also the side effect that the nearest timer may have been cancelled
> in-between and we might reprogram too-early but the event should be rare enough
> that we don't care.
>
> Another reason also is that cpuidle may have programmed a shallow C-state
> because it saw an early next expiration estimation. And if the related timer is
> cancelled in-between and we didn't keep the old expiration estimation, we would
> otherwise stop the tick for a long time with a shallow C-state.
I'll add a comment covering all your input! Thanks!
The probability that there happens a lot of enqueue and dequeue of
timers between get_next_timer_interrupt() and setting timer base idle is
not very high. But we have to make sure that we do not miss a new first
timer there.
>> @@ -926,7 +944,7 @@ static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
>> * first call we save the current tick time, so we can restart
>> * the scheduler tick in nohz_restart_sched_tick.
>> */
>> - if (!ts->tick_stopped) {
>> + if (!ts->tick_stopped && timer_idle) {
>
> In fact, if (!ts->tick_stopped && !timer_idle) then you
> should return now and avoid the reprogramming.
You are right. I'll add it and test it.
>> @@ -1950,6 +1950,40 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
>> if (cpu_is_offline(smp_processor_id()))
>> return expires;
>>
>> + raw_spin_lock(&base->lock);
>> + nextevt = __get_next_timer_interrupt(basej, base);
>> + raw_spin_unlock(&base->lock);
>
> It's unfortunate we have to lock here, which means we lock twice
> on the idle path. But I can't think of a better way and I guess
> the follow-up patches rely on that.
We have to do it like this, because power people need the sleep length
information to able to decide whether to stop the tick or not. If we do
not want to have the timer base locked two times in idle path, we will
not be able to move timer base idle marking into
tick_nohz_stop_tick(). But the good thing is, that we do not mark timer
bases idle, when tick is not stopped with this approach.
btw, I try to rewrite this patch completely as tglx was not happy about
some parts of code duplication. I'll make sure that your remarks are
also covered.
Thanks,
Anna-Maria
"Pandruvada, Srinivas" <[email protected]> writes:
> Hi Maria,
>
> On Wed, 2023-10-04 at 14:34 +0200, Anna-Maria Behnsen wrote:
>> Hi,
>>
>>
>
> [...]
>
>>
>> The proper solution to this problem is to always queue the timers on
>> the
>> local CPU and allow the non pinned timers to be pulled onto a busy
>> CPU at
>> expiry time.
>
> Thanks for these patches. I am looking for saving power during video
> playback with our low power daemon. I use cgroup v2 isolation to keep
> some CPUs idle (CPU 0-11) and video is played on a single module (CPU
> 12-15).
>
> I have some kernelshark pictures at below link. The traces are
> collected with sched, timer and irq. With 6.6-rc5, you can see some
> timers still expires on CPUs which I want to keep idle. With timer
> patches added, they are mostly pulled to busy CPU.
>
> https://imgur.com/a/8nF5OoP
>
> I can share the .dat files, but they are too big to attach here.
Thanks a lot for testing! The images are totally fine (at least for
me). As there are still some issues in v8, I'll have to post a new
version...
Thanks,
Anna-Maria
Hello Prateek,
K Prateek Nayak <[email protected]> writes:
> Hello Anna-Maria,
>
> Happy to report I don't see any regression with this version of series.
> I'll leave the detailed report below.
[...]
> Thank you for debugging and helping fix the tbench regression.
> If the series does not change drastically, feel free to add:
>
> Tested-by: K Prateek Nayak <[email protected]>
>
Thanks a lot for all the testing you did! When posting v9, I'll
summarize the changes and if required, I'll ask for testing support, if
it is ok?
>>
>> Possible Next Steps
>> ~~~~~~~~~~~~~~~~~~~
>>
>> Simple deferrable timers are no longer required as they can be converted to
>> global timers. If a CPU goes idle, a formerly deferrable timer will not
>> prevent the CPU to sleep as long as possible. Only the last migrator CPU
>> has to take care of them. Deferrable timers with timer pinned flags needs
>> to be expired on the specified CPU but must not prevent CPU from going
>> idle. They require their own timer base which is never taken into account
>> when calculating the next expiry time. This conversation and required
>> cleanup will be done in a follow up series.
>>
>
> I'll keep an eye out for future versions for testing.
I'll keep you in the loop.
Thanks,
Anna-Maria
Hi Lukasz,
Lukasz Luba <[email protected]> writes:
[...]
>
> I have tested this on my 2 Arm boards with mainline kernel
> and almost-mainline. On both platforms it looks stable.
> The results w/ your patchset looks better.
>
Thanks for testing!
[...]
> The performance looks good. Only one test 'Speedometer'
> has some interesting lower score.
Is it required to look into this more detailed or is the regression in a
acceptable range for you?
>
> Fill free to add:
>
> Tested-by: Lukasz Luba <[email protected]>
Thanks,
Anna-Maria
Hello Prateek,
I'm sorry for the late reply!
K Prateek Nayak <[email protected]> writes:
> Hello Anna-Maria,
>
> On 10/4/2023 6:04 PM, Anna-Maria Behnsen wrote:
>> [..snip..]
>>
>> Ping Pong Oberservation
>> ^^^^^^^^^^^^^^^^^^^^^^^
>>
>> During testing on a mostly idle machine a ping pong game could be observed:
>> a process_timeout timer is expired remotely on a non idle CPU. Then the CPU
>> where the schedule_timeout() was executed to enqueue the timer comes out of
>> idle and restarts the timer using schedule_timeout() and goes back to idle
>> again. This is due to the fair scheduler which tries to keep the task on
>> the CPU which it previously executed on.
>
> Regarding above, are you referring to "wake_up_process(timeout->task)" in
> "process_timeout()" ends up waking the task on an idle CPU instead of the
> CPU where process_timeout() ran?
Yes.
> In which case, have you tried using the "WF_CURRENT_CPU" flag for the
> wakeup? (landed upstream in v6.6-rc1) It is only used by wait queues in
> kernel/sched/wait.c currently but perhaps we can have a
> "wake_up_process_on_current_cpu()" that process_timeout() can call.
>
> Something along the lines of:
>
> int wake_up_process_on_current_cpu(struct task_struct *p)
> {
> return try_to_wake_up(p, TASK_NORMAL, WF_CURRENT_CPU);
> }
> EXPORT_SYMBOL(wake_up_process_on_current_cpu);
>
> Thoughts?
I didn't look into this again. Back than, I reported the observation to
scheduler people (others also already observed this behavior). I'm not
so familiar with scheduling, so I will ping scheduler people to give you
a feedback.
Thanks,
Anna-Maria
On 10/19/23 15:04, Anna-Maria Behnsen wrote:
> Hi Lukasz,
>
> Lukasz Luba <[email protected]> writes:
>
> [...]
>
>>
>> I have tested this on my 2 Arm boards with mainline kernel
>> and almost-mainline. On both platforms it looks stable.
>> The results w/ your patchset looks better.
>>
>
> Thanks for testing!
You're welcome
>
> [...]
>
>> The performance looks good. Only one test 'Speedometer'
>> has some interesting lower score.
>
> Is it required to look into this more detailed or is the regression in a
> acceptable range for you?
That's something which we can ignore. I have tested with a different
cpu idle governor and it goes away. So, I suspect that the governor
heuristic just probably was confused. I will spend some time on
different idle governor tuning.
Feel free to go forward with this patch set.
>
>>
>> Fill free to add:
>>
>> Tested-by: Lukasz Luba <[email protected]>
>
> Thanks,
>
> Anna-Maria
>
On Fri, Oct 06, 2023 at 10:35:43AM +0530, K Prateek Nayak wrote:
> Hello Anna-Maria,
>
> On 10/4/2023 6:04 PM, Anna-Maria Behnsen wrote:
> > [..snip..]
> >
> > Ping Pong Oberservation
> > ^^^^^^^^^^^^^^^^^^^^^^^
> >
> > During testing on a mostly idle machine a ping pong game could be observed:
> > a process_timeout timer is expired remotely on a non idle CPU. Then the CPU
> > where the schedule_timeout() was executed to enqueue the timer comes out of
> > idle and restarts the timer using schedule_timeout() and goes back to idle
> > again. This is due to the fair scheduler which tries to keep the task on
> > the CPU which it previously executed on.
>
> Regarding above, are you referring to "wake_up_process(timeout->task)" in
> "process_timeout()" ends up waking the task on an idle CPU instead of the
> CPU where process_timeout() ran?
>
> In which case, have you tried using the "WF_CURRENT_CPU" flag for the
> wakeup? (landed upstream in v6.6-rc1) It is only used by wait queues in
> kernel/sched/wait.c currently but perhaps we can have a
> "wake_up_process_on_current_cpu()" that process_timeout() can call.
>
> Something along the lines of:
>
> int wake_up_process_on_current_cpu(struct task_struct *p)
> {
> return try_to_wake_up(p, TASK_NORMAL, WF_CURRENT_CPU);
> }
> EXPORT_SYMBOL(wake_up_process_on_current_cpu);
>
> Thoughts?
Yeah, we should definitely not export such a function. Also,
WF_CURRENT_CPU should be used sparingly.
So restarting the task on the previous CPU is done because of cache
affinity and is typically the right thing to do if you care about
performance.
The first question to ask is where this schedule_timeout() is coming
from. Is this some daft userspace that is polling on something that
eventually ends up in schedule_timeout() ?
Can we fix the userspace to not do silly things like that?
The alternative is adding heuristics and we all know where that ends :/