Hi,
Patch 2 simply ignores any user passed slack time in the case of RT/DL
tasks for schedule_hrtimeout_range, which is aligned to what nanosleep(2)
already does.
Thanks!
Davidlohr Bueso (2):
kernel/hrtimer: Rely on rt_task() for DL tasks too
kernel/hrtimer: Ignore slack time for RT tasks
kernel/time/hrtimer.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
--
2.39.0
While in theory the timer can be triggered before expires+delta,
for the cases of RT tasks they really have no business giving
any lenience for extra slack time, so override any passed value
by the user and always use zero for schedule_hrtimeout_range()
calls. Furthermore, this is similar to what the nanosleep(2)
family already does with current->timer_slack_ns.
Signed-off-by: Davidlohr Bueso <[email protected]>
---
kernel/time/hrtimer.c | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 8336c2618ec1..78f2e07d3e7d 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -2270,7 +2270,7 @@ void __init hrtimers_init(void)
/**
* schedule_hrtimeout_range_clock - sleep until timeout
* @expires: timeout value (ktime_t)
- * @delta: slack in expires timeout (ktime_t)
+ * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks
* @mode: timer mode
* @clock_id: timer clock to be used
*/
@@ -2297,6 +2297,13 @@ schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta,
return -EINTR;
}
+ /*
+ * Override any slack passed by the user if under
+ * rt contraints.
+ */
+ if (rt_task(current))
+ delta = 0;
+
hrtimer_init_sleeper_on_stack(&t, clock_id, mode);
hrtimer_set_expires_range_ns(&t.timer, *expires, delta);
hrtimer_sleeper_start_expires(&t, mode);
@@ -2316,7 +2323,7 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock);
/**
* schedule_hrtimeout_range - sleep until timeout
* @expires: timeout value (ktime_t)
- * @delta: slack in expires timeout (ktime_t)
+ * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks
* @mode: timer mode
*
* Make the current task sleep until the given expiry time has
@@ -2324,7 +2331,8 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock);
* the current task state has been set (see set_current_state()).
*
* The @delta argument gives the kernel the freedom to schedule the
- * actual wakeup to a time that is both power and performance friendly.
+ * actual wakeup to a time that is both power and performance friendly
+ * for regular (non RT/DL) tasks.
* The kernel give the normal best effort behavior for "@expires+@delta",
* but may decide to fire the timer earlier, but no earlier than @expires.
*
--
2.39.0
The following commit has been merged into the timers/core branch of tip:
Commit-ID: 0c52310f260014d95c1310364379772cb74cf82d
Gitweb: https://git.kernel.org/tip/0c52310f260014d95c1310364379772cb74cf82d
Author: Davidlohr Bueso <[email protected]>
AuthorDate: Mon, 23 Jan 2023 09:32:06 -08:00
Committer: Thomas Gleixner <[email protected]>
CommitterDate: Tue, 31 Jan 2023 11:23:07 +01:00
hrtimer: Ignore slack time for RT tasks in schedule_hrtimeout_range()
While in theory the timer can be triggered before expires + delta, for the
cases of RT tasks they really have no business giving any lenience for
extra slack time, so override any passed value by the user and always use
zero for schedule_hrtimeout_range() calls. Furthermore, this is similar to
what the nanosleep(2) family already does with current->timer_slack_ns.
Signed-off-by: Davidlohr Bueso <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
kernel/time/hrtimer.c | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 667b713..e8c0829 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -2272,7 +2272,7 @@ void __init hrtimers_init(void)
/**
* schedule_hrtimeout_range_clock - sleep until timeout
* @expires: timeout value (ktime_t)
- * @delta: slack in expires timeout (ktime_t)
+ * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks
* @mode: timer mode
* @clock_id: timer clock to be used
*/
@@ -2299,6 +2299,13 @@ schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta,
return -EINTR;
}
+ /*
+ * Override any slack passed by the user if under
+ * rt contraints.
+ */
+ if (rt_task(current))
+ delta = 0;
+
hrtimer_init_sleeper_on_stack(&t, clock_id, mode);
hrtimer_set_expires_range_ns(&t.timer, *expires, delta);
hrtimer_sleeper_start_expires(&t, mode);
@@ -2318,7 +2325,7 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock);
/**
* schedule_hrtimeout_range - sleep until timeout
* @expires: timeout value (ktime_t)
- * @delta: slack in expires timeout (ktime_t)
+ * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks
* @mode: timer mode
*
* Make the current task sleep until the given expiry time has
@@ -2326,7 +2333,8 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock);
* the current task state has been set (see set_current_state()).
*
* The @delta argument gives the kernel the freedom to schedule the
- * actual wakeup to a time that is both power and performance friendly.
+ * actual wakeup to a time that is both power and performance friendly
+ * for regular (non RT/DL) tasks.
* The kernel give the normal best effort behavior for "@expires+@delta",
* but may decide to fire the timer earlier, but no earlier than @expires.
*