Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commitb4fb015

Browse files
koct9iIngo Molnar
authored and
Ingo Molnar
committed
sched/rt: Optimize checking group RT scheduler constraints
Group RT scheduler contains protection against setting zero runtime forcgroup with RT tasks. Right now function tg_set_rt_bandwidth() iteratesover all CPU cgroups and calls tg_has_rt_tasks() for any cgroup whichruntime is zero (not only for changed one). Default RT runtime is zero,thus tg_has_rt_tasks() will is called for almost at CPU cgroups.This protection already is slightly racy: runtime limit could be changedbetween cpu_cgroup_can_attach() and cpu_cgroup_attach() because changingcgroup attribute does not lock cgroup_mutex while attach does not lockrt_constraints_mutex. Changing task scheduler class also races withchanging rt runtime: check in __sched_setscheduler() isn't protected.Function tg_has_rt_tasks() iterates over all threads in the system.This gives NR_CGROUPS * NR_TASKS operations under single tasklist_locklocked for read tg_set_rt_bandwidth(). Any concurrent attempt of lockingtasklist_lock for write (for example fork) will stuck with disabled irqs.This patch makes two optimizations:1) Remove locking tasklist_lock and iterate only tasks in cgroup2) Call tg_has_rt_tasks() iff rt runtime changes from non-zero to zeroAll changed code is under CONFIG_RT_GROUP_SCHED.Testcase: # mkdir /sys/fs/cgroup/cpu/test{1..10000} # echo 0 | tee /sys/fs/cgroup/cpu/test*/cpu.rt_runtime_usAt the same time without patch fork time will be >100ms: # perf trace -e clone --duration 100 stress-ng --fork 1Also remote ping will show timings >100ms caused by irq latency.Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>Signed-off-by: Ingo Molnar <mingo@kernel.org>Link:https://lkml.kernel.org/r/157996383820.4651.11292439232549211693.stgit@buzz
1 parentbec2860 commitb4fb015

File tree

1 file changed

+11
-13
lines changed

1 file changed

+11
-13
lines changed

‎kernel/sched/rt.c‎

Lines changed: 11 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -2449,23 +2449,24 @@ const struct sched_class rt_sched_class = {
24492449
*/
24502450
staticDEFINE_MUTEX(rt_constraints_mutex);
24512451

2452-
/* Must be called with tasklist_lock held */
24532452
staticinlineinttg_has_rt_tasks(structtask_group*tg)
24542453
{
2455-
structtask_struct*g,*p;
2454+
structtask_struct*task;
2455+
structcss_task_iterit;
2456+
intret=0;
24562457

24572458
/*
24582459
* Autogroups do not have RT tasks; see autogroup_create().
24592460
*/
24602461
if (task_group_is_autogroup(tg))
24612462
return0;
24622463

2463-
for_each_process_thread(g,p) {
2464-
if (rt_task(p)&&task_group(p)==tg)
2465-
return1;
2466-
}
2464+
css_task_iter_start(&tg->css,0,&it);
2465+
while (!ret&&(task=css_task_iter_next(&it)))
2466+
ret |=rt_task(task);
2467+
css_task_iter_end(&it);
24672468

2468-
return0;
2469+
returnret;
24692470
}
24702471

24712472
structrt_schedulable_data {
@@ -2496,9 +2497,10 @@ static int tg_rt_schedulable(struct task_group *tg, void *data)
24962497
return-EINVAL;
24972498

24982499
/*
2499-
* Ensure we don't starve existing RT tasks.
2500+
* Ensure we don't starve existing RT tasks if runtime turns zero.
25002501
*/
2501-
if (rt_bandwidth_enabled()&& !runtime&&tg_has_rt_tasks(tg))
2502+
if (rt_bandwidth_enabled()&& !runtime&&
2503+
tg->rt_bandwidth.rt_runtime&&tg_has_rt_tasks(tg))
25022504
return-EBUSY;
25032505

25042506
total=to_ratio(period,runtime);
@@ -2564,7 +2566,6 @@ static int tg_set_rt_bandwidth(struct task_group *tg,
25642566
return-EINVAL;
25652567

25662568
mutex_lock(&rt_constraints_mutex);
2567-
read_lock(&tasklist_lock);
25682569
err=__rt_schedulable(tg,rt_period,rt_runtime);
25692570
if (err)
25702571
gotounlock;
@@ -2582,7 +2583,6 @@ static int tg_set_rt_bandwidth(struct task_group *tg,
25822583
}
25832584
raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock);
25842585
unlock:
2585-
read_unlock(&tasklist_lock);
25862586
mutex_unlock(&rt_constraints_mutex);
25872587

25882588
returnerr;
@@ -2641,9 +2641,7 @@ static int sched_rt_global_constraints(void)
26412641
intret=0;
26422642

26432643
mutex_lock(&rt_constraints_mutex);
2644-
read_lock(&tasklist_lock);
26452644
ret=__rt_schedulable(NULL,0,0);
2646-
read_unlock(&tasklist_lock);
26472645
mutex_unlock(&rt_constraints_mutex);
26482646

26492647
returnret;

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp