linux.git/kernel/sched, branch v4.19.192

sched/fair: Fix unfairness caused by missing load decay

2021-05-22T08:59:44+00:00

[ Upstream commit 0258bdfaff5bd13c4d2383150b7097aecd6b6d82 ]

This fixes an issue where old load on a cfs_rq is not properly decayed,
resulting in strange behavior where fairness can decrease drastically.
Real workloads with equally weighted control groups have ended up
getting a respective 99% and 1%(!!) of cpu time.

When an idle task is attached to a cfs_rq by attaching a pid to a cgroup,
the old load of the task is attached to the new cfs_rq and sched_entity by
attach_entity_cfs_rq. If the task is then moved to another cpu (and
therefore cfs_rq) before being enqueued/woken up, the load will be moved
to cfs_rq->removed from the sched_entity. Such a move will happen when
enforcing a cpuset on the task (eg. via a cgroup) that force it to move.

The load will however not be removed from the task_group itself, making
it look like there is a constant load on that cfs_rq. This causes the
vruntime of tasks on other sibling cfs_rq's to increase faster than they
are supposed to; causing severe fairness issues. If no other task is
started on the given cfs_rq, and due to the cpuset it would not happen,
this load would never be properly unloaded. With this patch the load
will be properly removed inside update_blocked_averages. This also
applies to tasks moved to the fair scheduling class and moved to another
cpu, and this path will also fix that. For fork, the entity is queued
right away, so this problem does not affect that.

This applies to cases where the new process is the first in the cfs_rq,
issue introduced 3d30544f0212 ("sched/fair: Apply more PELT fixes"), and
when there has previously been load on the cgroup but the cgroup was
removed from the leaflist due to having null PELT load, indroduced
in 039ae8bcf7a5 ("sched/fair: Fix O(nr_cgroups) in the load balancing
path").

For a simple cgroup hierarchy (as seen below) with two equally weighted
groups, that in theory should get 50/50 of cpu time each, it often leads
to a load of 60/40 or 70/30.

parent/
  cg-1/
    cpu.weight: 100
    cpuset.cpus: 1
  cg-2/
    cpu.weight: 100
    cpuset.cpus: 1

If the hierarchy is deeper (as seen below), while keeping cg-1 and cg-2
equally weighted, they should still get a 50/50 balance of cpu time.
This however sometimes results in a balance of 10/90 or 1/99(!!) between
the task groups.

$ ps u -C stress
USER         PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
root       18568  1.1  0.0   3684   100 pts/12   R+   13:36   0:00 stress --cpu 1
root       18580 99.3  0.0   3684   100 pts/12   R+   13:36   0:09 stress --cpu 1

parent/
  cg-1/
    cpu.weight: 100
    sub-group/
      cpu.weight: 1
      cpuset.cpus: 1
  cg-2/
    cpu.weight: 100
    sub-group/
      cpu.weight: 10000
      cpuset.cpus: 1

This can be reproduced by attaching an idle process to a cgroup and
moving it to a given cpuset before it wakes up. The issue is evident in
many (if not most) container runtimes, and has been reproduced
with both crun and runc (and therefore docker and all its "derivatives"),
and with both cgroup v1 and v2.

Fixes: 3d30544f0212 ("sched/fair: Apply more PELT fixes")
Fixes: 039ae8bcf7a5 ("sched/fair: Fix O(nr_cgroups) in the load balancing path")
Signed-off-by: Odin Ugedal 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Vincent Guittot 
Link: https://lkml.kernel.org/r/20210501141950.23622-2-odin@uged.al
Signed-off-by: Sasha Levin

sched/debug: Fix cgroup_path[] serialization

2021-05-22T08:59:34+00:00

[ Upstream commit ad789f84c9a145f8a18744c0387cec22ec51651e ]

The handling of sysrq key can be activated by echoing the key to
/proc/sysrq-trigger or via the magic key sequence typed into a terminal
that is connected to the system in some way (serial, USB or other mean).
In the former case, the handling is done in a user context. In the
latter case, it is likely to be in an interrupt context.

Currently in print_cpu() of kernel/sched/debug.c, sched_debug_lock is
taken with interrupt disabled for the whole duration of the calls to
print_*_stats() and print_rq() which could last for the quite some time
if the information dump happens on the serial console.

If the system has many cpus and the sched_debug_lock is somehow busy
(e.g. parallel sysrq-t), the system may hit a hard lockup panic
depending on the actually serial console implementation of the
system.

The purpose of sched_debug_lock is to serialize the use of the global
cgroup_path[] buffer in print_cpu(). The rests of the printk calls don't
need serialization from sched_debug_lock.

Calling printk() with interrupt disabled can still be problematic if
multiple instances are running. Allocating a stack buffer of PATH_MAX
bytes is not feasible because of the limited size of the kernel stack.

The solution implemented in this patch is to allow only one caller at a
time to use the full size group_path[], while other simultaneous callers
will have to use shorter stack buffers with the possibility of path
name truncation. A "..." suffix will be printed if truncation may have
happened.  The cgroup path name is provided for informational purpose
only, so occasional path name truncation should not be a big problem.

Fixes: efe25c2c7b3a ("sched: Reinstate group names in /proc/sched_debug")
Suggested-by: Peter Zijlstra 
Signed-off-by: Waiman Long 
Signed-off-by: Peter Zijlstra (Intel) 
Link: https://lkml.kernel.org/r/20210415195426.6677-1-longman@redhat.com
Signed-off-by: Sasha Levin

sched: Reenable interrupts in do_sched_yield()

2020-12-30T10:25:50+00:00

[ Upstream commit 345a957fcc95630bf5535d7668a59ed983eb49a7 ]

do_sched_yield() invokes schedule() with interrupts disabled which is
not allowed. This goes back to the pre git era to commit a6efb709806c
("[PATCH] irqlock patch 2.5.27-H6") in the history tree.

Reenable interrupts and remove the misleading comment which "explains" it.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Thomas Gleixner 
Signed-off-by: Peter Zijlstra (Intel) 
Link: https://lkml.kernel.org/r/87r1pt7y5c.fsf@nanos.tec.linutronix.de
Signed-off-by: Sasha Levin

sched/deadline: Fix sched_dl_global_validate()

2020-12-30T10:25:50+00:00

[ Upstream commit a57415f5d1e43c3a5c5d412cd85e2792d7ed9b11 ]

When change sched_rt_{runtime, period}_us, we validate that the new
settings should at least accommodate the currently allocated -dl
bandwidth:

  sched_rt_handler()
    -->	sched_dl_bandwidth_validate()
	{
		new_bw = global_rt_runtime()/global_rt_period();

		for_each_possible_cpu(cpu) {
			dl_b = dl_bw_of(cpu);
			if (new_bw < dl_b->total_bw)    <-------
				ret = -EBUSY;
		}
	}

But under CONFIG_SMP, dl_bw is per root domain , but not per CPU,
dl_b->total_bw is the allocated bandwidth of the whole root domain.
Instead, we should compare dl_b->total_bw against "cpus*new_bw",
where 'cpus' is the number of CPUs of the root domain.

Also, below annotation(in kernel/sched/sched.h) implied implementation
only appeared in SCHED_DEADLINE v2[1], then deadline scheduler kept
evolving till got merged(v9), but the annotation remains unchanged,
meaningless and misleading, update it.

* With respect to SMP, the bandwidth is given on a per-CPU basis,
* meaning that:
*  - dl_bw (< 100%) is the bandwidth of the system (group) on each CPU;
*  - dl_total_bw array contains, in the i-eth element, the currently
*    allocated bandwidth on the i-eth CPU.

[1]: https://lore.kernel.org/lkml/1267385230.13676.101.camel@Palantir/

Fixes: 332ac17ef5bf ("sched/deadline: Add bandwidth management for SCHED_DEADLINE tasks")
Signed-off-by: Peng Liu 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Daniel Bristot de Oliveira 
Acked-by: Juri Lelli 
Link: https://lkml.kernel.org/r/db6bbda316048cda7a1bbc9571defde193a8d67e.1602171061.git.iwtbavbm@gmail.com
Signed-off-by: Sasha Levin

sched/features: Fix !CONFIG_JUMP_LABEL case

2020-10-30T09:38:25+00:00

[ Upstream commit a73f863af4ce9730795eab7097fb2102e6854365 ]

Commit:

  765cc3a4b224e ("sched/core: Optimize sched_feat() for !CONFIG_SCHED_DEBUG builds")

made sched features static for !CONFIG_SCHED_DEBUG configurations, but
overlooked the CONFIG_SCHED_DEBUG=y and !CONFIG_JUMP_LABEL cases.

For the latter echoing changes to /sys/kernel/debug/sched_features has
the nasty effect of effectively changing what sched_features reports,
but without actually changing the scheduler behaviour (since different
translation units get different sysctl_sched_features).

Fix CONFIG_SCHED_DEBUG=y and !CONFIG_JUMP_LABEL configurations by properly
restructuring ifdefs.

Fixes: 765cc3a4b224e ("sched/core: Optimize sched_feat() for !CONFIG_SCHED_DEBUG builds")
Co-developed-by: Daniel Bristot de Oliveira 
Signed-off-by: Daniel Bristot de Oliveira 
Signed-off-by: Juri Lelli 
Signed-off-by: Ingo Molnar 
Acked-by: Patrick Bellasi 
Reviewed-by: Valentin Schneider 
Link: https://lore.kernel.org/r/20201013053114.160628-1-juri.lelli@redhat.com
Signed-off-by: Sasha Levin

sched: correct SD_flags returned by tl->sd_flags()

2020-08-19T06:14:47+00:00

[ Upstream commit 9b1b234bb86bcdcdb142e900d39b599185465dbb ]

During sched domain init, we check whether non-topological SD_flags are
returned by tl->sd_flags(), if found, fire a waning and correct the
violation, but the code failed to correct the violation. Correct this.

Fixes: 143e1e28cb40 ("sched: Rework sched_domain topology definition")
Signed-off-by: Peng Liu 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Vincent Guittot 
Reviewed-by: Valentin Schneider 
Link: https://lkml.kernel.org/r/20200609150936.GA13060@iZj6chx1xj0e0buvshuecpZ
Signed-off-by: Sasha Levin

sched/fair: Fix NOHZ next idle balance

2020-08-19T06:14:47+00:00

[ Upstream commit 3ea2f097b17e13a8280f1f9386c331b326a3dbef ]

With commit:
  'b7031a02ec75 ("sched/fair: Add NOHZ_STATS_KICK")'
rebalance_domains of the local cfs_rq happens before others idle cpus have
updated nohz.next_balance and its value is overwritten.

Move the update of nohz.next_balance for other idles cpus before balancing
and updating the next_balance of local cfs_rq.

Also, the nohz.next_balance is now updated only if all idle cpus got a
chance to rebalance their domains and the idle balance has not been aborted
because of new activities on the CPU. In case of need_resched, the idle
load balance will be kick the next jiffie in order to address remaining
ilb.

Fixes: b7031a02ec75 ("sched/fair: Add NOHZ_STATS_KICK")
Reported-by: Peng Liu 
Signed-off-by: Vincent Guittot 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Valentin Schneider 
Acked-by: Mel Gorman 
Link: https://lkml.kernel.org/r/20200609123748.18636-1-vincent.guittot@linaro.org
Signed-off-by: Sasha Levin

sched/fair: handle case of task_h_load() returning 0

2020-07-22T07:32:13+00:00

commit 01cfcde9c26d8555f0e6e9aea9d6049f87683998 upstream.

task_h_load() can return 0 in some situations like running stress-ng
mmapfork, which forks thousands of threads, in a sched group on a 224 cores
system. The load balance doesn't handle this correctly because
env->imbalance never decreases and it will stop pulling tasks only after
reaching loop_max, which can be equal to the number of running tasks of
the cfs. Make sure that imbalance will be decreased by at least 1.

misfit task is the other feature that doesn't handle correctly such
situation although it's probably more difficult to face the problem
because of the smaller number of CPUs and running tasks on heterogenous
system.

We can't simply ensure that task_h_load() returns at least one because it
would imply to handle underflow in other places.

Signed-off-by: Vincent Guittot 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Valentin Schneider 
Reviewed-by: Dietmar Eggemann 
Tested-by: Dietmar Eggemann 
Cc:  # v4.4+
Link: https://lkml.kernel.org/r/20200710152426.16981-1-vincent.guittot@linaro.org
Signed-off-by: Greg Kroah-Hartman

sched: Fix unreliable rseq cpu_id for new tasks

2020-07-22T07:32:13+00:00

commit ce3614daabea8a2d01c1dd17ae41d1ec5e5ae7db upstream.

While integrating rseq into glibc and replacing glibc's sched_getcpu
implementation with rseq, glibc's tests discovered an issue with
incorrect __rseq_abi.cpu_id field value right after the first time
a newly created process issues sched_setaffinity.

For the records, it triggers after building glibc and running tests, and
then issuing:

  for x in {1..2000} ; do posix/tst-affinity-static  & done

and shows up as:

error: Unexpected CPU 2, expected 0
error: Unexpected CPU 2, expected 0
error: Unexpected CPU 2, expected 0
error: Unexpected CPU 2, expected 0
error: Unexpected CPU 138, expected 0
error: Unexpected CPU 138, expected 0
error: Unexpected CPU 138, expected 0
error: Unexpected CPU 138, expected 0

This is caused by the scheduler invoking __set_task_cpu() directly from
sched_fork() and wake_up_new_task(), thus bypassing rseq_migrate() which
is done by set_task_cpu().

Add the missing rseq_migrate() to both functions. The only other direct
use of __set_task_cpu() is done by init_idle(), which does not involve a
user-space task.

Based on my testing with the glibc test-case, just adding rseq_migrate()
to wake_up_new_task() is sufficient to fix the observed issue. Also add
it to sched_fork() to keep things consistent.

The reason why this never triggered so far with the rseq/basic_test
selftest is unclear.

The current use of sched_getcpu(3) does not typically require it to be
always accurate. However, use of the __rseq_abi.cpu_id field within rseq
critical sections requires it to be accurate. If it is not accurate, it
can cause corruption in the per-cpu data targeted by rseq critical
sections in user-space.

Reported-By: Florian Weimer 
Signed-off-by: Mathieu Desnoyers 
Signed-off-by: Peter Zijlstra (Intel) 
Tested-By: Florian Weimer 
Cc: stable@vger.kernel.org # v4.18+
Link: https://lkml.kernel.org/r/20200707201505.2632-1-mathieu.desnoyers@efficios.com
Signed-off-by: Greg Kroah-Hartman

sched/core: Fix PI boosting between RT and DEADLINE tasks

2020-07-01T03:17:13+00:00

[ Upstream commit 740797ce3a124b7dd22b7fb832d87bc8fba1cf6f ]

syzbot reported the following warning:

 WARNING: CPU: 1 PID: 6351 at kernel/sched/deadline.c:628
 enqueue_task_dl+0x22da/0x38a0 kernel/sched/deadline.c:1504

At deadline.c:628 we have:

 623 static inline void setup_new_dl_entity(struct sched_dl_entity *dl_se)
 624 {
 625 	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
 626 	struct rq *rq = rq_of_dl_rq(dl_rq);
 627
 628 	WARN_ON(dl_se->dl_boosted);
 629 	WARN_ON(dl_time_before(rq_clock(rq), dl_se->deadline));
        [...]
     }

Which means that setup_new_dl_entity() has been called on a task
currently boosted. This shouldn't happen though, as setup_new_dl_entity()
is only called when the 'dynamic' deadline of the new entity
is in the past w.r.t. rq_clock and boosted tasks shouldn't verify this
condition.

Digging through the PI code I noticed that what above might in fact happen
if an RT tasks blocks on an rt_mutex hold by a DEADLINE task. In the
first branch of boosting conditions we check only if a pi_task 'dynamic'
deadline is earlier than mutex holder's and in this case we set mutex
holder to be dl_boosted. However, since RT 'dynamic' deadlines are only
initialized if such tasks get boosted at some point (or if they become
DEADLINE of course), in general RT 'dynamic' deadlines are usually equal
to 0 and this verifies the aforementioned condition.

Fix it by checking that the potential donor task is actually (even if
temporary because in turn boosted) running at DEADLINE priority before
using its 'dynamic' deadline value.

Fixes: 2d3d891d3344 ("sched/deadline: Add SCHED_DEADLINE inheritance logic")
Reported-by: syzbot+119ba87189432ead09b4@syzkaller.appspotmail.com
Signed-off-by: Juri Lelli 
Signed-off-by: Peter Zijlstra (Intel) 
Signed-off-by: Ingo Molnar 
Reviewed-by: Daniel Bristot de Oliveira 
Tested-by: Daniel Wagner 
Link: https://lkml.kernel.org/r/20181119153201.GB2119@localhost.localdomain
Signed-off-by: Sasha Levin