With the advent of various new memory types, some machines will have
multiple types of memory, e.g. DRAM and PMEM (persistent memory). The
memory subsystem of these machines can be called memory tiering system,
because the performance of the different types of memory are usually
different.
In such system, because of the memory accessing pattern changing etc,
some pages in the slow memory may become hot globally. So in this
patch, the NUMA balancing mechanism is enhanced to optimize the page
placement among the different memory types according to hot/cold
dynamically.
In a typical memory tiering system, there are CPUs, fast memory and slow
memory in each physical NUMA node. The CPUs and the fast memory will be
put in one logical node (called fast memory node), while the slow memory
will be put in another (faked) logical node (called slow memory node).
That is, the fast memory is regarded as local while the slow memory is
regarded as remote. So it's possible for the recently accessed pages in
the slow memory node to be promoted to the fast memory node via the
existing NUMA balancing mechanism.
The original NUMA balancing mechanism will stop to migrate pages if the
free memory of the target node becomes below the high watermark. This
is a reasonable policy if there's only one memory type. But this makes
the original NUMA balancing mechanism almost do not work to optimize
page placement among different memory types. Details are as follows.
It's the common cases that the working-set size of the workload is
larger than the size of the fast memory nodes. Otherwise, it's
unnecessary to use the slow memory at all. So, there are almost always
no enough free pages in the fast memory nodes, so that the globally hot
pages in the slow memory node cannot be promoted to the fast memory
node. To solve the issue, we have 2 choices as follows,
a. Ignore the free pages watermark checking when promoting hot pages
from the slow memory node to the fast memory node. This will
create some memory pressure in the fast memory node, thus trigger
the memory reclaiming. So that, the cold pages in the fast memory
node will be demoted to the slow memory node.
b. Define a new watermark called wmark_promo which is higher than
wmark_high, and have kswapd reclaiming pages until free pages reach
such watermark. The scenario is as follows: when we want to promote
hot-pages from a slow memory to a fast memory, but fast memory's free
pages would go lower than high watermark with such promotion, we wake
up kswapd with wmark_promo watermark in order to demote cold pages and
free us up some space. So, next time we want to promote hot-pages we
might have a chance of doing so.
The choice "a" may create high memory pressure in the fast memory node.
If the memory pressure of the workload is high, the memory pressure
may become so high that the memory allocation latency of the workload
is influenced, e.g. the direct reclaiming may be triggered.
The choice "b" works much better at this aspect. If the memory
pressure of the workload is high, the hot pages promotion will stop
earlier because its allocation watermark is higher than that of the
normal memory allocation. So in this patch, choice "b" is implemented.
A new zone watermark (WMARK_PROMO) is added. Which is larger than the
high watermark and can be controlled via watermark_scale_factor.
In addition to the original page placement optimization among sockets,
the NUMA balancing mechanism is extended to be used to optimize page
placement according to hot/cold among different memory types. So the
sysctl user space interface (numa_balancing) is extended in a backward
compatible way as follow, so that the users can enable/disable these
functionality individually.
The sysctl is converted from a Boolean value to a bits field. The
definition of the flags is,
- 0: NUMA_BALANCING_DISABLED
- 1: NUMA_BALANCING_NORMAL
- 2: NUMA_BALANCING_MEMORY_TIERING
We have tested the patch with the pmbench memory accessing benchmark
with the 80:20 read/write ratio and the Gauss access address
distribution on a 2 socket Intel server with Optane DC Persistent
Memory Model. The test results shows that the pmbench score can
improve up to 95.9%.
Thanks Andrew Morton to help fix the document format error.
Link: https://lkml.kernel.org/r/20220221084529.1052339-3-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Feng Tang <feng.tang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Where commit 4ef0c5c6b5 ("kernel/sched: Fix sched_fork() access an
invalid sched_task_group") fixed a fork race vs cgroup, it opened up a
race vs syscalls by not placing the task on the runqueue before it
gets exposed through the pidhash.
Commit 13765de814 ("sched/fair: Fix fault in reweight_entity") is
trying to fix a single instance of this, instead fix the whole class
of issues, effectively reverting this commit.
Fixes: 4ef0c5c6b5 ("kernel/sched: Fix sched_fork() access an invalid sched_task_group")
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tadeusz Struk <tadeusz.struk@linaro.org>
Tested-by: Zhang Qiao <zhangqiao22@huawei.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lkml.kernel.org/r/YgoeCbwj5mbCR0qA@hirez.programming.kicks-ass.net
Syzbot found a GPF in reweight_entity. This has been bisected to
commit 4ef0c5c6b5 ("kernel/sched: Fix sched_fork() access an invalid
sched_task_group")
There is a race between sched_post_fork() and setpriority(PRIO_PGRP)
within a thread group that causes a null-ptr-deref in
reweight_entity() in CFS. The scenario is that the main process spawns
number of new threads, which then call setpriority(PRIO_PGRP, 0, -20),
wait, and exit. For each of the new threads the copy_process() gets
invoked, which adds the new task_struct and calls sched_post_fork()
for it.
In the above scenario there is a possibility that
setpriority(PRIO_PGRP) and set_one_prio() will be called for a thread
in the group that is just being created by copy_process(), and for
which the sched_post_fork() has not been executed yet. This will
trigger a null pointer dereference in reweight_entity(), as it will
try to access the run queue pointer, which hasn't been set.
Before the mentioned change the cfs_rq pointer for the task has been
set in sched_fork(), which is called much earlier in copy_process(),
before the new task is added to the thread_group. Now it is done in
the sched_post_fork(), which is called after that. To fix the issue
the remove the update_load param from the update_load param() function
and call reweight_task() only if the task flag doesn't have the
TASK_NEW flag set.
Fixes: 4ef0c5c6b5 ("kernel/sched: Fix sched_fork() access an invalid sched_task_group")
Reported-by: syzbot+af7a719bc92395ee41b3@syzkaller.appspotmail.com
Signed-off-by: Tadeusz Struk <tadeusz.struk@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20220203161846.1160750-1-tadeusz.struk@linaro.org
set when querying membarrier(2) commands through MEMBARRIER_CMD_QUERY
-----BEGIN PGP SIGNATURE-----
iQIzBAABCgAdFiEEzv7L6UO9uDPlPSfHEsHwGGHeVUoFAmH2cDYACgkQEsHwGGHe
VUo7YhAAs4trrOaq5KCr2YW7agAHqYjYsMfPLEBUOvJ4hGggMAYCwGXYDdzMo8UH
tAxLfh2fyy3VUNKPeLMqz6XVn697oTQnSAkVBDJIcMJlctZGvXmwNnv3g133xlML
n4xLhVAfC76HA0Q0zkPZ6gdN/VBrRl94n3mAWxB9FVFUDwFIMr6PkGrjw9Olml/r
Xm2WUuKfvwHqEhNUDu2rS0qqal3mPO6DV+4Y7JGyQL5fYAu3HfabZV4CBfYt5z3P
m3y4Y2yfpWcB4D6KfXHPQwLhWOoxVZ0X2YiAjfj8x/+dYaTIoBSg/EG+44nX6tBV
RcYK9gARwfCgSdcfWZFYYfHlGbNv1x/HHOVvkuJEzfCya4+FQTRdrqkoOWsPUPk1
jqa9Ybz9wQxJ2GI57wT7W+fyl2M+agyvwywrELLy6w6nwAKdWpbIGW4b1ummfIk2
MGqL99Hge6aX9ONFT6IA4rHd0fvWNC00l4evzNVfyHQnYN3f0ul8h8p950F09I25
apSalhyz40LbJdKsRFGt8CjIgM2Y+rP87JI8ZawjauWFIp+lLiXcLOuLXTNJjAe6
Sw9EWkkr3sTHzOxudrFHz/QwM+m7KkoYnuGQw0gDqzXdqc1Gpc91llLqqWNuyX++
y3NWzEdmnjlV+H1Fls94UhrdsHAiI8d+OTH3fQVY6VEtGcLQJ2c=
=kI2Z
-----END PGP SIGNATURE-----
Merge tag 'sched_urgent_for_v5.17_rc2_p2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fix from Borislav Petkov:
"Make sure the membarrier-rseq fence commands are part of the reported
set when querying membarrier(2) commands through MEMBARRIER_CMD_QUERY"
* tag 'sched_urgent_for_v5.17_rc2_p2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/membarrier: Fix membarrier-rseq fence command missing from query bitmask
When CONFIG_PROC_FS is disabled psi code generates the following
warnings:
kernel/sched/psi.c:1364:30: warning: 'psi_cpu_proc_ops' defined but not used [-Wunused-const-variable=]
1364 | static const struct proc_ops psi_cpu_proc_ops = {
| ^~~~~~~~~~~~~~~~
kernel/sched/psi.c:1355:30: warning: 'psi_memory_proc_ops' defined but not used [-Wunused-const-variable=]
1355 | static const struct proc_ops psi_memory_proc_ops = {
| ^~~~~~~~~~~~~~~~~~~
kernel/sched/psi.c:1346:30: warning: 'psi_io_proc_ops' defined but not used [-Wunused-const-variable=]
1346 | static const struct proc_ops psi_io_proc_ops = {
| ^~~~~~~~~~~~~~~
Make definitions of these structures and related functions conditional
on CONFIG_PROC_FS config.
Link: https://lkml.kernel.org/r/20220119223940.787748-3-surenb@google.com
Fixes: 0e94682b73 ("psi: introduce psi monitor")
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reported-by: kernel test robot <lkp@intel.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The membarrier command MEMBARRIER_CMD_QUERY allows querying the
available membarrier commands. When the membarrier-rseq fence commands
were added, a new MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ_BITMASK was
introduced with the intent to expose them with the MEMBARRIER_CMD_QUERY
command, the but it was never added to MEMBARRIER_CMD_BITMASK.
The membarrier-rseq fence commands are therefore not wired up with the
query command.
Rename MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ_BITMASK to
MEMBARRIER_PRIVATE_EXPEDITED_RSEQ_BITMASK (the bitmask is not a command
per-se), and change the erroneous
MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_RSEQ_BITMASK (which does not
actually exist) to MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_RSEQ.
Wire up MEMBARRIER_PRIVATE_EXPEDITED_RSEQ_BITMASK in
MEMBARRIER_CMD_BITMASK. Fixing this allows discovering availability of
the membarrier-rseq fence feature.
Fixes: 2a36ab717e ("rseq/membarrier: Add MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ")
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org> # 5.10+
Link: https://lkml.kernel.org/r/20220117203010.30129-1-mathieu.desnoyers@efficios.com
propagation in the sched hierarchies and other minor cleanups and
improvements
-----BEGIN PGP SIGNATURE-----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=4m7x
-----END PGP SIGNATURE-----
Merge tag 'sched_urgent_for_v5.17_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Borislav Petkov:
"A bunch of fixes: forced idle time accounting, utilization values
propagation in the sched hierarchies and other minor cleanups and
improvements"
* tag 'sched_urgent_for_v5.17_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
kernel/sched: Remove dl_boosted flag comment
sched: Avoid double preemption in __cond_resched_*lock*()
sched/fair: Fix all kernel-doc warnings
sched/core: Accounting forceidle time for all tasks except idle task
sched/pelt: Relax the sync of load_sum with load_avg
sched/pelt: Relax the sync of runnable_sum with runnable_avg
sched/pelt: Continue to relax the sync of util_sum with util_avg
sched/pelt: Relax the sync of util_sum with util_avg
psi: Fix uaf issue when psi trigger is destroyed while being polled
For PREEMPT/DYNAMIC_PREEMPT the *_unlock() will already trigger a
preemption, no point in then calling preempt_schedule_common()
*again*.
Use _cond_resched() instead, since this is a NOP for the preemptible
configs while it provide a preemption point for the others.
Reported-by: xuhaifeng <xuhaifeng@oppo.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/YcGnvDEYBwOiV0cR@hirez.programming.kicks-ass.net
Quieten all kernel-doc warnings in kernel/sched/fair.c:
kernel/sched/fair.c:3663: warning: No description found for return value of 'update_cfs_rq_load_avg'
kernel/sched/fair.c:8601: warning: No description found for return value of 'asym_smt_can_pull_tasks'
kernel/sched/fair.c:8673: warning: Function parameter or member 'sds' not described in 'update_sg_lb_stats'
kernel/sched/fair.c:9483: warning: contents before sections
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20211218055900.2704-1-rdunlap@infradead.org
There are two types of forced idle time: forced idle time from cookie'd
task and forced idle time form uncookie'd task. The forced idle time from
uncookie'd task is actually caused by the cookie'd task in runqueue
indirectly, and it's more accurate to measure the capacity loss with the
sum of both.
Assuming cpu x and cpu y are a pair of SMT siblings, consider the
following scenarios:
1.There's a cookie'd task running on cpu x, and there're 4 uncookie'd
tasks running on cpu y. For cpu x, there will be 80% forced idle time
(from uncookie'd task); for cpu y, there will be 20% forced idle time
(from cookie'd task).
2.There's a uncookie'd task running on cpu x, and there're 4 cookie'd
tasks running on cpu y. For cpu x, there will be 80% forced idle time
(from cookie'd task); for cpu y, there will be 20% forced idle time
(from uncookie'd task).
The scenario1 can recurrent by stress-ng(scenario2 can recurrent similary):
(cookie'd)taskset -c x stress-ng -c 1 -l 100
(uncookie'd)taskset -c y stress-ng -c 4 -l 100
In the above two scenarios, the total capacity loss is 1 cpu, but in
scenario1, the cookie'd forced idle time tells us 20% cpu capacity loss, in
scenario2, the cookie'd forced idle time tells us 80% cpu capacity loss,
which are not accurate. It'll be more accurate to measure with cookie'd
forced idle time and uncookie'd forced idle time.
Signed-off-by: Cruz Zhao <CruzZhao@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Don <joshdon@google.com>
Link: https://lore.kernel.org/r/1641894961-9241-2-git-send-email-CruzZhao@linux.alibaba.com
Similarly to util_avg and util_sum, don't sync load_sum with the low
bound of load_avg but only ensure that load_sum stays in the correct range.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-5-vincent.guittot@linaro.org
Similarly to util_avg and util_sum, don't sync runnable_sum with the low
bound of runnable_avg but only ensure that runnable_sum stays in the
correct range.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-4-vincent.guittot@linaro.org
Rick reported performance regressions in bugzilla because of cpu frequency
being lower than before:
https://bugzilla.kernel.org/show_bug.cgi?id=215045
He bisected the problem to:
commit 1c35b07e6d ("sched/fair: Ensure _sum and _avg values stay consistent")
This commit forces util_sum to be synced with the new util_avg after
removing the contribution of a task and before the next periodic sync. By
doing so util_sum is rounded to its lower bound and might lost up to
LOAD_AVG_MAX-1 of accumulated contribution which has not yet been
reflected in util_avg.
update_tg_cfs_util() is not the only place where we round util_sum and
lost some accumulated contributions that are not already reflected in
util_avg. Modify update_tg_cfs_util() and detach_entity_load_avg() to not
sync util_sum with the new util_avg. Instead of always setting util_sum to
the low bound of util_avg, which can significantly lower the utilization,
we propagate the difference. In addition, we also check that cfs's util_sum
always stays above the lower bound for a given util_avg as it has been
observed that sched_entity's util_sum is sometimes above cfs one.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-3-vincent.guittot@linaro.org
Rick reported performance regressions in bugzilla because of cpu frequency
being lower than before:
https://bugzilla.kernel.org/show_bug.cgi?id=215045
He bisected the problem to:
commit 1c35b07e6d ("sched/fair: Ensure _sum and _avg values stay consistent")
This commit forces util_sum to be synced with the new util_avg after
removing the contribution of a task and before the next periodic sync. By
doing so util_sum is rounded to its lower bound and might lost up to
LOAD_AVG_MAX-1 of accumulated contribution which has not yet been
reflected in util_avg.
Instead of always setting util_sum to the low bound of util_avg, which can
significantly lower the utilization of root cfs_rq after propagating the
change down into the hierarchy, we revert the change of util_sum and
propagate the difference.
In addition, we also check that cfs's util_sum always stays above the
lower bound for a given util_avg as it has been observed that
sched_entity's util_sum is sometimes above cfs one.
Fixes: 1c35b07e6d ("sched/fair: Ensure _sum and _avg values stay consistent")
Reported-by: Rick Yiu <rickyiu@google.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-2-vincent.guittot@linaro.org
With write operation on psi files replacing old trigger with a new one,
the lifetime of its waitqueue is totally arbitrary. Overwriting an
existing trigger causes its waitqueue to be freed and pending poll()
will stumble on trigger->event_wait which was destroyed.
Fix this by disallowing to redefine an existing psi trigger. If a write
operation is used on a file descriptor with an already existing psi
trigger, the operation will fail with EBUSY error.
Also bypass a check for psi_disabled in the psi_trigger_destroy as the
flag can be flipped after the trigger is created, leading to a memory
leak.
Fixes: 0e94682b73 ("psi: introduce psi monitor")
Reported-by: syzbot+cdb5dd11c97cc532efad@syzkaller.appspotmail.com
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Analyzed-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20220111232309.1786347-1-surenb@google.com
Pull signal/exit/ptrace updates from Eric Biederman:
"This set of changes deletes some dead code, makes a lot of cleanups
which hopefully make the code easier to follow, and fixes bugs found
along the way.
The end-game which I have not yet reached yet is for fatal signals
that generate coredumps to be short-circuit deliverable from
complete_signal, for force_siginfo_to_task not to require changing
userspace configured signal delivery state, and for the ptrace stops
to always happen in locations where we can guarantee on all
architectures that the all of the registers are saved and available on
the stack.
Removal of profile_task_ext, profile_munmap, and profile_handoff_task
are the big successes for dead code removal this round.
A bunch of small bug fixes are included, as most of the issues
reported were small enough that they would not affect bisection so I
simply added the fixes and did not fold the fixes into the changes
they were fixing.
There was a bug that broke coredumps piped to systemd-coredump. I
dropped the change that caused that bug and replaced it entirely with
something much more restrained. Unfortunately that required some
rebasing.
Some successes after this set of changes: There are few enough calls
to do_exit to audit in a reasonable amount of time. The lifetime of
struct kthread now matches the lifetime of struct task, and the
pointer to struct kthread is no longer stored in set_child_tid. The
flag SIGNAL_GROUP_COREDUMP is removed. The field group_exit_task is
removed. Issues where task->exit_code was examined with
signal->group_exit_code should been examined were fixed.
There are several loosely related changes included because I am
cleaning up and if I don't include them they will probably get lost.
The original postings of these changes can be found at:
https://lkml.kernel.org/r/87a6ha4zsd.fsf@email.froward.int.ebiederm.orghttps://lkml.kernel.org/r/87bl1kunjj.fsf@email.froward.int.ebiederm.orghttps://lkml.kernel.org/r/87r19opkx1.fsf_-_@email.froward.int.ebiederm.org
I trimmed back the last set of changes to only the obviously correct
once. Simply because there was less time for review than I had hoped"
* 'signal-for-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (44 commits)
ptrace/m68k: Stop open coding ptrace_report_syscall
ptrace: Remove unused regs argument from ptrace_report_syscall
ptrace: Remove second setting of PT_SEIZED in ptrace_attach
taskstats: Cleanup the use of task->exit_code
exit: Use the correct exit_code in /proc/<pid>/stat
exit: Fix the exit_code for wait_task_zombie
exit: Coredumps reach do_group_exit
exit: Remove profile_handoff_task
exit: Remove profile_task_exit & profile_munmap
signal: clean up kernel-doc comments
signal: Remove the helper signal_group_exit
signal: Rename group_exit_task group_exec_task
coredump: Stop setting signal->group_exit_task
signal: Remove SIGNAL_GROUP_COREDUMP
signal: During coredumps set SIGNAL_GROUP_EXIT in zap_process
signal: Make coredump handling explicit in complete_signal
signal: Have prepare_signal detect coredumps using signal->core_state
signal: Have the oom killer detect coredumps using signal->core_state
exit: Move force_uaccess back into do_exit
exit: Guarantee make_task_dead leaks the tsk when calling do_task_exit
...
"Lots of cleanups and preparation; highlights:
- futex: Cleanup and remove runtime futex_cmpxchg detection
- rtmutex: Some fixes for the PREEMPT_RT locking infrastructure
- kcsan: Share owner_on_cpu() between mutex,rtmutex and rwsem and
annotate the racy owner->on_cpu access *once*.
- atomic64: Dead-Code-Elemination"
-----BEGIN PGP SIGNATURE-----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=vrYy
-----END PGP SIGNATURE-----
Merge tag 'locking_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking updates from Borislav Petkov:
"Lots of cleanups and preparation. Highlights:
- futex: Cleanup and remove runtime futex_cmpxchg detection
- rtmutex: Some fixes for the PREEMPT_RT locking infrastructure
- kcsan: Share owner_on_cpu() between mutex,rtmutex and rwsem and
annotate the racy owner->on_cpu access *once*.
- atomic64: Dead-Code-Elemination"
[ Description above by Peter Zijlstra ]
* tag 'locking_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/atomic: atomic64: Remove unusable atomic ops
futex: Fix additional regressions
locking: Allow to include asm/spinlock_types.h from linux/spinlock_types_raw.h
x86/mm: Include spinlock_t definition in pgtable.
locking: Mark racy reads of owner->on_cpu
locking: Make owner_on_cpu() into <linux/sched.h>
lockdep/selftests: Adapt ww-tests for PREEMPT_RT
lockdep/selftests: Skip the softirq related tests on PREEMPT_RT
lockdep/selftests: Unbalanced migrate_disable() & rcu_read_lock().
lockdep/selftests: Avoid using local_lock_{acquire|release}().
lockdep: Remove softirq accounting on PREEMPT_RT.
locking/rtmutex: Add rt_mutex_lock_nest_lock() and rt_mutex_lock_killable().
locking/rtmutex: Squash self-deadlock check for ww_rt_mutex.
locking: Remove rt_rwlock_is_contended().
sched: Trigger warning if ->migration_disabled counter underflows.
futex: Fix sparc32/m68k/nds32 build regression
futex: Remove futex_cmpxchg detection
futex: Ensure futex_atomic_cmpxchg_inatomic() is present
kernel/locking: Use a pointer in ww_mutex_trylock().
"Mostly minor things this time; some highlights:
- core-sched: Add 'Forced Idle' accounting; this allows to track how
much CPU time is 'lost' due to core scheduling constraints.
- psi: Fix for MEM_FULL; a task running reclaim would be counted as a
runnable task and prevent MEM_FULL from being reported.
- cpuacct: Long standing fixes for some cgroup accounting issues.
- rt: Bandwidth timer could, under unusual circumstances, be failed to
armed, leading to indefinite throttling."
-----BEGIN PGP SIGNATURE-----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=H0vR
-----END PGP SIGNATURE-----
Merge tag 'sched_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler updates from Borislav Petkov:
"Mostly minor things this time; some highlights:
- core-sched: Add 'Forced Idle' accounting; this allows to track how
much CPU time is 'lost' due to core scheduling constraints.
- psi: Fix for MEM_FULL; a task running reclaim would be counted as a
runnable task and prevent MEM_FULL from being reported.
- cpuacct: Long standing fixes for some cgroup accounting issues.
- rt: Bandwidth timer could, under unusual circumstances, be failed
to armed, leading to indefinite throttling."
[ Description above by Peter Zijlstra ]
* tag 'sched_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Replace CFS internal cpu_util() with cpu_util_cfs()
sched/fair: Cleanup task_util and capacity type
sched/rt: Try to restart rt period timer when rt runtime exceeded
sched/fair: Document the slow path and fast path in select_task_rq_fair
sched/fair: Fix per-CPU kthread and wakee stacking for asym CPU capacity
sched/fair: Fix detection of per-CPU kthreads waking a task
sched/cpuacct: Make user/system times in cpuacct.stat more precise
sched/cpuacct: Fix user/system in shown cpuacct.usage*
cpuacct: Convert BUG_ON() to WARN_ON_ONCE()
cputime, cpuacct: Include guest time in user time in cpuacct.stat
psi: Fix PSI_MEM_FULL state when tasks are in memstall and doing reclaim
sched/core: Forced idle accounting
psi: Add a missing SPDX license header
psi: Remove repeated verbose comment
This series provides KCSAN fixes and also the ability to take memory
barriers into account for weakly-ordered systems. This last can increase
the probability of detecting certain types of data races.
-----BEGIN PGP SIGNATURE-----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=Q5fg
-----END PGP SIGNATURE-----
Merge tag 'kcsan.2022.01.09a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu
Pull KCSAN updates from Paul McKenney:
"This provides KCSAN fixes and also the ability to take memory barriers
into account for weakly-ordered systems. This last can increase the
probability of detecting certain types of data races"
* tag 'kcsan.2022.01.09a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu: (29 commits)
kcsan: Only test clear_bit_unlock_is_negative_byte if arch defines it
kcsan: Avoid nested contexts reading inconsistent reorder_access
kcsan: Turn barrier instrumentation into macros
kcsan: Make barrier tests compatible with lockdep
kcsan: Support WEAK_MEMORY with Clang where no objtool support exists
compiler_attributes.h: Add __disable_sanitizer_instrumentation
objtool, kcsan: Remove memory barrier instrumentation from noinstr
objtool, kcsan: Add memory barrier instrumentation to whitelist
sched, kcsan: Enable memory barrier instrumentation
mm, kcsan: Enable barrier instrumentation
x86/qspinlock, kcsan: Instrument barrier of pv_queued_spin_unlock()
x86/barriers, kcsan: Use generic instrumentation for non-smp barriers
asm-generic/bitops, kcsan: Add instrumentation for barriers
locking/atomics, kcsan: Add instrumentation for barriers
locking/barriers, kcsan: Support generic instrumentation
locking/barriers, kcsan: Add instrumentation for barriers
kcsan: selftest: Add test case to check memory barrier instrumentation
kcsan: Ignore GCC 11+ warnings about TSan runtime support
kcsan: test: Add test cases for memory barrier instrumentation
kcsan: test: Match reordered or normal accesses
...
accesing it in order to prevent any potential data races, and convert
all users to those new accessors
-----BEGIN PGP SIGNATURE-----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=v6Xn
-----END PGP SIGNATURE-----
Merge tag 'core_entry_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull thread_info flag accessor helper updates from Borislav Petkov:
"Add a set of thread_info.flags accessors which snapshot it before
accesing it in order to prevent any potential data races, and convert
all users to those new accessors"
* tag 'core_entry_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
powerpc: Snapshot thread flags
powerpc: Avoid discarding flags in system_call_exception()
openrisc: Snapshot thread flags
microblaze: Snapshot thread flags
arm64: Snapshot thread flags
ARM: Snapshot thread flags
alpha: Snapshot thread flags
sched: Snapshot thread flags
entry: Snapshot thread flags
x86: Snapshot thread flags
thread_info: Add helpers to snapshot thread flags
The point of using set_child_tid to hold the kthread pointer was that
it already did what is necessary. There are now restrictions on when
set_child_tid can be initialized and when set_child_tid can be used in
schedule_tail. Which indicates that continuing to use set_child_tid
to hold the kthread pointer is a bad idea.
Instead of continuing to use the set_child_tid field of task_struct
generalize the pf_io_worker field of task_struct and use it to hold
the kthread pointer.
Rename pf_io_worker (which is a void * pointer) to worker_private so
it can be used to store kthreads struct kthread pointer. Update the
kthread code to store the kthread pointer in the worker_private field.
Remove the places where set_child_tid had to be dealt with carefully
because kthreads also used it.
Link: https://lkml.kernel.org/r/CAHk-=wgtFAA9SbVYg0gR1tqPMC17-NYcs0GQkaYg1bGhh1uJQQ@mail.gmail.com
Link: https://lkml.kernel.org/r/87a6grvqy8.fsf_-_@email.froward.int.ebiederm.org
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Kernel threads abuse set_child_tid. Historically that has been fine
as set_child_tid was initialized after the kernel thread had been
forked. Unfortunately storing struct kthread in set_child_tid after
the thread is running makes struct kthread being unusable for storing
result codes of the thread.
When set_child_tid is set to struct kthread during fork that results
in schedule_tail writing the thread id to the beggining of struct
kthread (if put_user does not realize it is a kernel address).
Solve this by skipping the put_user for all kthreads.
Reported-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lkml.kernel.org/r/YcNsG0Lp94V13whH@archlinux-ax161
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Today the rules are a bit iffy and arbitrary about which kernel
threads have struct kthread present. Both idle threads and thread
started with create_kthread want struct kthread present so that is
effectively all kernel threads. Make the rule that if PF_KTHREAD
and the task is running then struct kthread is present.
This will allow the kernel thread code to using tsk->exit_code
with different semantics from ordinary processes.
To make ensure that struct kthread is present for all
kernel threads move it's allocation into copy_process.
Add a deallocation of struct kthread in exec for processes
that were kernel threads.
Move the allocation of struct kthread for the initial thread
earlier so that it is not repeated for each additional idle
thread.
Move the initialization of struct kthread into set_kthread_struct
so that the structure is always and reliably initailized.
Clear set_child_tid in free_kthread_struct to ensure the kthread
struct is reliably freed during exec. The function
free_kthread_struct does not need to clear vfork_done during exec as
exec_mm_release called from exec_mmap has already cleared vfork_done.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
cpu_util_cfs() was created by commit d4edd662ac ("sched/cpufreq: Use
the DEADLINE utilization signal") to enable the access to CPU
utilization from the Schedutil CPUfreq governor.
Commit a07630b8b2 ("sched/cpufreq/schedutil: Use util_est for OPP
selection") added util_est support later.
The only thing cpu_util() is doing on top of what cpu_util_cfs() already
does is to clamp the return value to the [0..capacity_orig] capacity
range of the CPU. Integrating this into cpu_util_cfs() is not harming
the existing users (Schedutil and CPUfreq cooling (latter via
sched_cpu_util() wrapper)).
For straightforwardness, prefer to keep using `int cpu` as the function
parameter over using `struct rq *rq` which might avoid some calls to
cpu_rq(cpu) -> per_cpu(runqueues, cpu) -> RELOC_HIDE().
Update cfs_util()'s documentation and reuse it for cpu_util_cfs().
Remove cpu_util().
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20211118164240.623551-1-dietmar.eggemann@arm.com
There's no fundamental reason to disable KCSAN for scheduler code,
except for excessive noise and performance concerns (instrumenting
scheduler code is usually a good way to stress test KCSAN itself).
However, several core sched functions imply memory barriers that are
invisible to KCSAN without instrumentation, but are required to avoid
false positives. Therefore, unconditionally enable instrumentation of
memory barriers in scheduler code. Also update the comment to reflect
this and be a bit more brief.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Several ->poll() implementations are special in that they use a
waitqueue whose lifetime is the current task, rather than the struct
file as is normally the case. This is okay for blocking polls, since a
blocking poll occurs within one task; however, non-blocking polls
require another solution. This solution is for the queue to be cleared
before it is freed, using 'wake_up_poll(wq, EPOLLHUP | POLLFREE);'.
However, that has a bug: wake_up_poll() calls __wake_up() with
nr_exclusive=1. Therefore, if there are multiple "exclusive" waiters,
and the wakeup function for the first one returns a positive value, only
that one will be called. That's *not* what's needed for POLLFREE;
POLLFREE is special in that it really needs to wake up everyone.
Considering the three non-blocking poll systems:
- io_uring poll doesn't handle POLLFREE at all, so it is broken anyway.
- aio poll is unaffected, since it doesn't support exclusive waits.
However, that's fragile, as someone could add this feature later.
- epoll doesn't appear to be broken by this, since its wakeup function
returns 0 when it sees POLLFREE. But this is fragile.
Although there is a workaround (see epoll), it's better to define a
function which always sends POLLFREE to all waiters. Add such a
function. Also make it verify that the queue really becomes empty after
all waiters have been woken up.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20211209010455.42744-2-ebiggers@kernel.org
Signed-off-by: Eric Biggers <ebiggers@google.com>
task_util and capacity are comparable unsigned long values. There is no
need for an intermidiate implicit signed cast.
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20211207095755.859972-1-vincent.donnefort@arm.com
When rt_runtime is modified from -1 to a valid control value, it may
cause the task to be throttled all the time. Operations like the following
will trigger the bug. E.g:
1. echo -1 > /proc/sys/kernel/sched_rt_runtime_us
2. Run a FIFO task named A that executes while(1)
3. echo 950000 > /proc/sys/kernel/sched_rt_runtime_us
When rt_runtime is -1, The rt period timer will not be activated when task
A enqueued. And then the task will be throttled after setting rt_runtime to
950,000. The task will always be throttled because the rt period timer is
not activated.
Fixes: d0b27fa778 ("sched: rt-group: synchonised bandwidth period")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Li Hua <hucool.lihua@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20211203033618.11895-1-hucool.lihua@huawei.com
All People I know including myself took a long time to figure out that
typical wakeup will always go to fast path and never go to slow path
except WF_FORK and WF_EXEC.
Vincent reminded me once in a linaro meeting and made me understand
slow path won't happen for WF_TTWU. But my other friends repeatedly
wasted a lot of time on testing this path like me before I reminded
them.
So obviously the code needs some document.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20211016111109.5559-1-21cnbao@gmail.com
If migrate_enable() is used more often than its counter part then it
remains undetected and rq::nr_pinned will underflow, too.
Add a warning if migrate_enable() is attempted if without a matching a
migrate_disable().
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211129174654.668506-2-bigeasy@linutronix.de
select_idle_sibling() has a special case for tasks woken up by a per-CPU
kthread where the selected CPU is the previous one. For asymmetric CPU
capacity systems, the assumption was that the wakee couldn't have a
bigger utilization during task placement than it used to have during the
last activation. That was not considering uclamp.min which can completely
change between two task activations and as a consequence mandates the
fitness criterion asym_fits_capacity(), even for the exit path described
above.
Fixes: b4c9c9f156 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lkml.kernel.org/r/20211129173115.4006346-1-vincent.donnefort@arm.com
select_idle_sibling() has a special case for tasks woken up by a per-CPU
kthread, where the selected CPU is the previous one. However, the current
condition for this exit path is incomplete. A task can wake up from an
interrupt context (e.g. hrtimer), while a per-CPU kthread is running. A
such scenario would spuriously trigger the special case described above.
Also, a recent change made the idle task like a regular per-CPU kthread,
hence making that situation more likely to happen
(is_per_cpu_kthread(swapper) being true now).
Checking for task context makes sure select_idle_sibling() will not
interpret a wake up from any other context as a wake up by a per-CPU
kthread.
Fixes: 52262ee567 ("sched/fair: Allow a per-CPU kthread waking a task to stack on the same CPU, to fix XFS performance regression")
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lore.kernel.org/r/20211201143450.479472-1-vincent.donnefort@arm.com
Commit d81ae8aac8 ("sched/uclamp: Fix initialization of struct
uclamp_rq") introduced a bug where uclamp_max of the rq is not reset to
match the woken up task's uclamp_max when the rq is idle.
The code was relying on rq->uclamp_max initialized to zero, so on first
enqueue
static inline void uclamp_rq_inc_id(struct rq *rq, struct task_struct *p,
enum uclamp_id clamp_id)
{
...
if (uc_se->value > READ_ONCE(uc_rq->value))
WRITE_ONCE(uc_rq->value, uc_se->value);
}
was actually resetting it. But since commit d81ae8aac8 changed the
default to 1024, this no longer works. And since rq->uclamp_flags is
also initialized to 0, neither above code path nor uclamp_idle_reset()
update the rq->uclamp_max on first wake up from idle.
This is only visible from first wake up(s) until the first dequeue to
idle after enabling the static key. And it only matters if the
uclamp_max of this task is < 1024 since only then its uclamp_max will be
effectively ignored.
Fix it by properly initializing rq->uclamp_flags = UCLAMP_FLAG_IDLE to
ensure uclamp_idle_reset() is called which then will update the rq
uclamp_max value as expected.
Fixes: d81ae8aac8 ("sched/uclamp: Fix initialization of struct uclamp_rq")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <Valentin.Schneider@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lkml.kernel.org/r/20211202112033.1705279-1-qais.yousef@arm.com
__setup() callbacks expect 1 for success and 0 for failure. Correct the
usage here to reflect that.
Fixes: 826bfeb37b ("preempt/dynamic: Support dynamic preempt with preempt= boot option")
Reported-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20211203233203.133581-1-ahalaney@redhat.com
getrusage(RUSAGE_THREAD) with nohz_full may return shorter utime/stime
than the actual time.
task_cputime_adjusted() snapshots utime and stime and then adjust their
sum to match the scheduler maintained cputime.sum_exec_runtime.
Unfortunately in nohz_full, sum_exec_runtime is only updated once per
second in the worst case, causing a discrepancy against utime and stime
that can be updated anytime by the reader using vtime.
To fix this situation, perform an update of cputime.sum_exec_runtime
when the cputime snapshot reports the task as actually running while
the tick is disabled. The related overhead is then contained within the
relevant situations.
Reported-by: Hasegawa Hitomi <hasegawa-hitomi@fujitsu.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Hasegawa Hitomi <hasegawa-hitomi@fujitsu.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Acked-by: Phil Auld <pauld@redhat.com>
Link: https://lore.kernel.org/r/20211026141055.57358-3-frederic@kernel.org
Some thread flags can be set remotely, and so even when IRQs are disabled,
the flags can change under our feet. Generally this is unlikely to cause a
problem in practice, but it is somewhat unsound, and KCSAN will
legitimately warn that there is a data race.
To avoid such issues, a snapshot of the flags has to be taken prior to
using them. Some places already use READ_ONCE() for that, others do not.
Convert them all to the new flag accessor helpers.
The READ_ONCE(ti->flags) .. cmpxchg(ti->flags) loop in
set_nr_if_polling() is left as-is for clarity.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20211129130653.2037928-4-mark.rutland@arm.com
To hot unplug a CPU, the idle task on that CPU calls a few layers of C
code before finally leaving the kernel. When KASAN is in use, poisoned
shadow is left around for each of the active stack frames, and when
shadow call stacks are in use. When shadow call stacks (SCS) are in use
the task's saved SCS SP is left pointing at an arbitrary point within
the task's shadow call stack.
When a CPU is offlined than onlined back into the kernel, this stale
state can adversely affect execution. Stale KASAN shadow can alias new
stackframes and result in bogus KASAN warnings. A stale SCS SP is
effectively a memory leak, and prevents a portion of the shadow call
stack being used. Across a number of hotplug cycles the idle task's
entire shadow call stack can become unusable.
We previously fixed the KASAN issue in commit:
e1b77c9298 ("sched/kasan: remove stale KASAN poison after hotplug")
... by removing any stale KASAN stack poison immediately prior to
onlining a CPU.
Subsequently in commit:
f1a0a376ca ("sched/core: Initialize the idle task with preemption disabled")
... the refactoring left the KASAN and SCS cleanup in one-time idle
thread initialization code rather than something invoked prior to each
CPU being onlined, breaking both as above.
We fixed SCS (but not KASAN) in commit:
63acd42c0d ("sched/scs: Reset the shadow stack when idle_task_exit")
... but as this runs in the context of the idle task being offlined it's
potentially fragile.
To fix these consistently and more robustly, reset the SCS SP and KASAN
shadow of a CPU's idle task immediately before we online that CPU in
bringup_cpu(). This ensures the idle task always has a consistent state
when it is running, and removes the need to so so when exiting an idle
task.
Whenever any thread is created, dup_task_struct() will give the task a
stack which is free of KASAN shadow, and initialize the task's SCS SP,
so there's no need to specially initialize either for idle thread within
init_idle(), as this was only necessary to handle hotplug cycles.
I've tested this on arm64 with:
* gcc 11.1.0, defconfig +KASAN_INLINE, KASAN_STACK
* clang 12.0.0, defconfig +KASAN_INLINE, KASAN_STACK, SHADOW_CALL_STACK
... offlining and onlining CPUS with:
| while true; do
| for C in /sys/devices/system/cpu/cpu*/online; do
| echo 0 > $C;
| echo 1 > $C;
| done
| done
Fixes: f1a0a376ca ("sched/core: Initialize the idle task with preemption disabled")
Reported-by: Qian Cai <quic_qiancai@quicinc.com>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Tested-by: Qian Cai <quic_qiancai@quicinc.com>
Link: https://lore.kernel.org/lkml/20211115113310.35693-1-mark.rutland@arm.com/
cpuacct.stat shows user time based on raw random precision tick
based counters. Use cputime_addjust() to scale these values against the
total runtime accounted by the scheduler, like we already do
for user/system times in /proc/<pid>/stat.
Signed-off-by: Andrey Ryabinin <arbn@yandex-team.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20211115164607.23784-4-arbn@yandex-team.com
cpuacct has 2 different ways of accounting and showing user
and system times.
The first one uses cpuacct_account_field() to account times
and cpuacct.stat file to expose them. And this one seems to work ok.
The second one is uses cpuacct_charge() function for accounting and
set of cpuacct.usage* files to show times. Despite some attempts to
fix it in the past it still doesn't work. Sometimes while running KVM
guest the cpuacct_charge() accounts most of the guest time as
system time. This doesn't match with user&system times shown in
cpuacct.stat or proc/<pid>/stat.
Demonstration:
# git clone https://github.com/aryabinin/kvmsample
# make
# mkdir /sys/fs/cgroup/cpuacct/test
# echo $$ > /sys/fs/cgroup/cpuacct/test/tasks
# ./kvmsample &
# for i in {1..5}; do cat /sys/fs/cgroup/cpuacct/test/cpuacct.usage_sys; sleep 1; done
1976535645
2979839428
3979832704
4983603153
5983604157
Use cpustats accounted in cpuacct_account_field() as the source
of user/sys times for cpuacct.usage* files. Make cpuacct_charge()
to account only summary execution time.
Fixes: d740037fac ("sched/cpuacct: Split usage accounting into user_usage and sys_usage")
Signed-off-by: Andrey Ryabinin <arbn@yandex-team.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20211115164607.23784-3-arbn@yandex-team.com
Replace fatal BUG_ON() with more safe WARN_ON_ONCE() in cpuacct_cpuusage_read().
Signed-off-by: Andrey Ryabinin <arbn@yandex-team.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20211115164607.23784-2-arbn@yandex-team.com
cpuacct.stat in no-root cgroups shows user time without guest time
included int it. This doesn't match with user time shown in root
cpuacct.stat and /proc/<pid>/stat. This also affects cgroup2's cpu.stat
in the same way.
Make account_guest_time() to add user time to cgroup's cpustat to
fix this.
Fixes: ef12fefabf ("cpuacct: add per-cgroup utime/stime statistics")
Signed-off-by: Andrey Ryabinin <arbn@yandex-team.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20211115164607.23784-1-arbn@yandex-team.com
We've noticed cases where tasks in a cgroup are stalled on memory but
there is little memory FULL pressure since tasks stay on the runqueue
in reclaim.
A simple example involves a single threaded program that keeps leaking
and touching large amounts of memory. It runs in a cgroup with swap
enabled, memory.high set at 10M and cpu.max ratio set at 5%. Though
there is significant CPU pressure and memory SOME, there is barely any
memory FULL since the task enters reclaim and stays on the runqueue.
However, this memory-bound task is effectively stalled on memory and
we expect memory FULL to match memory SOME in this scenario.
The code is confused about memstall && running, thinking there is a
stalled task and a productive task when there's only one task: a
reclaimer that's counted as both. To fix this, we redefine the
condition for PSI_MEM_FULL to check that all running tasks are in an
active memstall instead of checking that there are no running tasks.
case PSI_MEM_FULL:
- return unlikely(tasks[NR_MEMSTALL] && !tasks[NR_RUNNING]);
+ return unlikely(tasks[NR_MEMSTALL] &&
+ tasks[NR_RUNNING] == tasks[NR_MEMSTALL_RUNNING]);
This will capture reclaimers. It will also capture tasks that called
psi_memstall_enter() and are about to sleep, but this should be
negligible noise.
Signed-off-by: Brian Chen <brianchen118@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Link: https://lore.kernel.org/r/20211110213312.310243-1-brianchen118@gmail.com
Adds accounting for "forced idle" time, which is time where a cookie'd
task forces its SMT sibling to idle, despite the presence of runnable
tasks.
Forced idle time is one means to measure the cost of enabling core
scheduling (ie. the capacity lost due to the need to force idle).
Forced idle time is attributed to the thread responsible for causing
the forced idle.
A few details:
- Forced idle time is displayed via /proc/PID/sched. It also requires
that schedstats is enabled.
- Forced idle is only accounted when a sibling hyperthread is held
idle despite the presence of runnable tasks. No time is charged if
a sibling is idle but has no runnable tasks.
- Tasks with 0 cookie are never charged forced idle.
- For SMT > 2, we scale the amount of forced idle charged based on the
number of forced idle siblings. Additionally, we split the time up and
evenly charge it to all running tasks, as each is equally responsible
for the forced idle.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20211018203428.2025792-1-joshdon@google.com
Add the missing SPDX license header to
include/linux/psi.h
include/linux/psi_types.h
kernel/sched/psi.c
Signed-off-by: Liu Xinpeng <liuxp11@chinatelecom.cn>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Link: https://lore.kernel.org/r/1635133586-84611-2-git-send-email-liuxp11@chinatelecom.cn
Comment in function psi_task_switch,there are two same lines.
...
* runtime state, the cgroup that contains both tasks
* runtime state, the cgroup that contains both tasks
...
Signed-off-by: Liu Xinpeng <liuxp11@chinatelecom.cn>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Link: https://lore.kernel.org/r/1635133586-84611-1-git-send-email-liuxp11@chinatelecom.cn
Commit c597bfddc9 ("sched: Provide Kconfig support for default dynamic
preempt mode") changed the selectable config names for the preemption
model. This means a config file must now select
CONFIG_PREEMPT_BEHAVIOUR=y
rather than
CONFIG_PREEMPT=y
to get a preemptible kernel. This means all arch config files would need to
be updated - right now they'll all end up with the default
CONFIG_PREEMPT_NONE_BEHAVIOUR.
Rather than touch a good hundred of config files, restore usage of
CONFIG_PREEMPT{_NONE, _VOLUNTARY}. Make them configure:
o The build-time preemption model when !PREEMPT_DYNAMIC
o The default boot-time preemption model when PREEMPT_DYNAMIC
Add siblings of those configs with the _BUILD suffix to unconditionally
designate the build-time preemption model (PREEMPT_DYNAMIC is built with
the "highest" preemption model it supports, aka PREEMPT). Downstream
configs should by now all be depending / selected by CONFIG_PREEMPTION
rather than CONFIG_PREEMPT, so only a few sites need patching up.
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Marco Elver <elver@google.com>
Link: https://lore.kernel.org/r/20211110202448.4054153-2-valentin.schneider@arm.com
Kevin is reporting crashes which point to a use-after-free of a cfs_rq
in update_blocked_averages(). Initial debugging revealed that we've
live cfs_rq's (on_list=1) in an about to be kfree()'d task group in
free_fair_sched_group(). However, it was unclear how that can happen.
His kernel config happened to lead to a layout of struct sched_entity
that put the 'my_q' member directly into the middle of the object
which makes it incidentally overlap with SLUB's freelist pointer.
That, in combination with SLAB_FREELIST_HARDENED's freelist pointer
mangling, leads to a reliable access violation in form of a #GP which
made the UAF fail fast.
Michal seems to have run into the same issue[1]. He already correctly
diagnosed that commit a7b359fc6a ("sched/fair: Correctly insert
cfs_rq's to list on unthrottle") is causing the preconditions for the
UAF to happen by re-adding cfs_rq's also to task groups that have no
more running tasks, i.e. also to dead ones. His analysis, however,
misses the real root cause and it cannot be seen from the crash
backtrace only, as the real offender is tg_unthrottle_up() getting
called via sched_cfs_period_timer() via the timer interrupt at an
inconvenient time.
When unregister_fair_sched_group() unlinks all cfs_rq's from the dying
task group, it doesn't protect itself from getting interrupted. If the
timer interrupt triggers while we iterate over all CPUs or after
unregister_fair_sched_group() has finished but prior to unlinking the
task group, sched_cfs_period_timer() will execute and walk the list of
task groups, trying to unthrottle cfs_rq's, i.e. re-add them to the
dying task group. These will later -- in free_fair_sched_group() -- be
kfree()'ed while still being linked, leading to the fireworks Kevin
and Michal are seeing.
To fix this race, ensure the dying task group gets unlinked first.
However, simply switching the order of unregistering and unlinking the
task group isn't sufficient, as concurrent RCU walkers might still see
it, as can be seen below:
CPU1: CPU2:
: timer IRQ:
: do_sched_cfs_period_timer():
: :
: distribute_cfs_runtime():
: rcu_read_lock();
: :
: unthrottle_cfs_rq():
sched_offline_group(): :
: walk_tg_tree_from(…,tg_unthrottle_up,…):
list_del_rcu(&tg->list); :
(1) : list_for_each_entry_rcu(child, &parent->children, siblings)
: :
(2) list_del_rcu(&tg->siblings); :
: tg_unthrottle_up():
unregister_fair_sched_group(): struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)];
: :
list_del_leaf_cfs_rq(tg->cfs_rq[cpu]); :
: :
: if (!cfs_rq_is_decayed(cfs_rq) || cfs_rq->nr_running)
(3) : list_add_leaf_cfs_rq(cfs_rq);
: :
: :
: :
: :
: :
(4) : rcu_read_unlock();
CPU 2 walks the task group list in parallel to sched_offline_group(),
specifically, it'll read the soon to be unlinked task group entry at
(1). Unlinking it on CPU 1 at (2) therefore won't prevent CPU 2 from
still passing it on to tg_unthrottle_up(). CPU 1 now tries to unlink
all cfs_rq's via list_del_leaf_cfs_rq() in
unregister_fair_sched_group(). Meanwhile CPU 2 will re-add some of
these at (3), which is the cause of the UAF later on.
To prevent this additional race from happening, we need to wait until
walk_tg_tree_from() has finished traversing the task groups, i.e.
after the RCU read critical section ends in (4). Afterwards we're safe
to call unregister_fair_sched_group(), as each new walk won't see the
dying task group any more.
On top of that, we need to wait yet another RCU grace period after
unregister_fair_sched_group() to ensure print_cfs_stats(), which might
run concurrently, always sees valid objects, i.e. not already free'd
ones.
This patch survives Michal's reproducer[2] for 8h+ now, which used to
trigger within minutes before.
[1] https://lore.kernel.org/lkml/20211011172236.11223-1-mkoutny@suse.com/
[2] https://lore.kernel.org/lkml/20211102160228.GA57072@blackbody.suse.cz/
Fixes: a7b359fc6a ("sched/fair: Correctly insert cfs_rq's to list on unthrottle")
[peterz: shuffle code around a bit]
Reported-by: Kevin Tanguy <kevin.tanguy@corp.ovh.com>
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Nothing protects the access to the per_cpu variable sd_llc_id. When testing
the same CPU (i.e. this_cpu == that_cpu), a race condition exists with
update_top_cache_domain(). One scenario being:
CPU1 CPU2
==================================================================
per_cpu(sd_llc_id, CPUX) => 0
partition_sched_domains_locked()
detach_destroy_domains()
cpus_share_cache(CPUX, CPUX) update_top_cache_domain(CPUX)
per_cpu(sd_llc_id, CPUX) => 0
per_cpu(sd_llc_id, CPUX) = CPUX
per_cpu(sd_llc_id, CPUX) => CPUX
return false
ttwu_queue_cond() wouldn't catch smp_processor_id() == cpu and the result
is a warning triggered from ttwu_queue_wakelist().
Avoid a such race in cpus_share_cache() by always returning true when
this_cpu == that_cpu.
Fixes: 518cd62341 ("sched: Only queue remote wakeups when crossing cache boundaries")
Reported-by: Jing-Ting Wu <jing-ting.wu@mediatek.com>
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20211104175120.857087-1-vincent.donnefort@arm.com
Huang Ying
Peter Zijlstra
Tadeusz Struk
Linus Torvalds
Suren Baghdasaryan
Mathieu Desnoyers
Randy Dunlap
Cruz Zhao
Vincent Guittot
Eric W. Biederman
Ingo Molnar
Dietmar Eggemann
Marco Elver
Eric Biggers
Vincent Donnefort
Li Hua
Barry Song
Sebastian Andrzej Siewior
Qais Yousef
Andrew Halaney
Frederic Weisbecker
Mark Rutland
Andrey Ryabinin
Brian Chen
Josh Don
Liu Xinpeng
Valentin Schneider
Mathias Krause