Bringing up a CPU may involve creating and destroying tasks which requires
read-locking threadgroup_rwsem, so threadgroup_rwsem nests inside
cpus_read_lock(). However, cpuset's ->attach(), which may be called with
thredagroup_rwsem write-locked, also wants to disable CPU hotplug and
acquires cpus_read_lock(), leading to a deadlock.
Fix it by guaranteeing that ->attach() is always called with CPU hotplug
disabled and removing cpus_read_lock() call from cpuset_attach().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-and-tested-by: Imran Khan <imran.f.khan@oracle.com>
Reported-and-tested-by: Xuewen Yan <xuewen.yan@unisoc.com>
Fixes: 05c7b7a92c ("cgroup/cpuset: Fix a race between cpuset_attach() and cpu hotplug")
Cc: stable@vger.kernel.org # v5.17+
Pull scheduler fixes from Ingo Molnar:
"Various fixes: a deadline scheduler fix, a migration fix, a Sparse fix
and a comment fix"
* tag 'sched-urgent-2022-08-06' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/core: Do not requeue task on CPU excluded from cpus_mask
sched/rt: Fix Sparse warnings due to undefined rt.c declarations
exit: Fix typo in comment: s/sub-theads/sub-threads
sched, cpuset: Fix dl_cpu_busy() panic due to empty cs->cpus_allowed
Pull cgroup updates from Tejun Heo:
"Several core optimizations:
- threadgroup_rwsem write locking is skipped when configuring
controllers in empty subtrees.
Combined with CLONE_INTO_CGROUP, this allows the common static
usage pattern to not grab threadgroup_rwsem at all (glibc still
doesn't seem ready for CLONE_INTO_CGROUP unfortunately).
- threadgroup_rwsem used to be put into non-percpu mode by default
due to latency concerns in specific use cases. There's no reason
for everyone else to pay for it. Make the behavior optional.
- psi no longer allocates memory when disabled.
... along with some code cleanups"
* tag 'cgroup-for-5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup: Skip subtree root in cgroup_update_dfl_csses()
cgroup: remove "no" prefixed mount options
cgroup: Make !percpu threadgroup_rwsem operations optional
cgroup: Add "no" prefixed mount options
cgroup: Elide write-locking threadgroup_rwsem when updating csses on an empty subtree
cgroup.c: remove redundant check for mixable cgroup in cgroup_migrate_vet_dst
cgroup.c: add helper __cset_cgroup_from_root to cleanup duplicated codes
psi: dont alloc memory for psi by default
With cgroup v2, the cpuset's cpus_allowed mask can be empty indicating
that the cpuset will just use the effective CPUs of its parent. So
cpuset_can_attach() can call task_can_attach() with an empty mask.
This can lead to cpumask_any_and() returns nr_cpu_ids causing the call
to dl_bw_of() to crash due to percpu value access of an out of bound
CPU value. For example:
[80468.182258] BUG: unable to handle page fault for address: ffffffff8b6648b0
:
[80468.191019] RIP: 0010:dl_cpu_busy+0x30/0x2b0
:
[80468.207946] Call Trace:
[80468.208947] cpuset_can_attach+0xa0/0x140
[80468.209953] cgroup_migrate_execute+0x8c/0x490
[80468.210931] cgroup_update_dfl_csses+0x254/0x270
[80468.211898] cgroup_subtree_control_write+0x322/0x400
[80468.212854] kernfs_fop_write_iter+0x11c/0x1b0
[80468.213777] new_sync_write+0x11f/0x1b0
[80468.214689] vfs_write+0x1eb/0x280
[80468.215592] ksys_write+0x5f/0xe0
[80468.216463] do_syscall_64+0x5c/0x80
[80468.224287] entry_SYSCALL_64_after_hwframe+0x44/0xae
Fix that by using effective_cpus instead. For cgroup v1, effective_cpus
is the same as cpus_allowed. For v2, effective_cpus is the real cpumask
to be used by tasks within the cpuset anyway.
Also update task_can_attach()'s 2nd argument name to cs_effective_cpus to
reflect the change. In addition, a check is added to task_can_attach()
to guard against the possibility that cpumask_any_and() may return a
value >= nr_cpu_ids.
Fixes: 7f51412a41 ("sched/deadline: Fix bandwidth check/update when migrating tasks between exclusive cpusets")
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20220803015451.2219567-1-longman@redhat.com
Pull scheduler updates from Ingo Molnar:
"Load-balancing improvements:
- Improve NUMA balancing on AMD Zen systems for affine workloads.
- Improve the handling of reduced-capacity CPUs in load-balancing.
- Energy Model improvements: fix & refine all the energy fairness
metrics (PELT), and remove the conservative threshold requiring 6%
energy savings to migrate a task. Doing this improves power
efficiency for most workloads, and also increases the reliability
of energy-efficiency scheduling.
- Optimize/tweak select_idle_cpu() to spend (much) less time
searching for an idle CPU on overloaded systems. There's reports of
several milliseconds spent there on large systems with large
workloads ...
[ Since the search logic changed, there might be behavioral side
effects. ]
- Improve NUMA imbalance behavior. On certain systems with spare
capacity, initial placement of tasks is non-deterministic, and such
an artificial placement imbalance can persist for a long time,
hurting (and sometimes helping) performance.
The fix is to make fork-time task placement consistent with runtime
NUMA balancing placement.
Note that some performance regressions were reported against this,
caused by workloads that are not memory bandwith limited, which
benefit from the artificial locality of the placement bug(s). Mel
Gorman's conclusion, with which we concur, was that consistency is
better than random workload benefits from non-deterministic bugs:
"Given there is no crystal ball and it's a tradeoff, I think
it's better to be consistent and use similar logic at both fork
time and runtime even if it doesn't have universal benefit."
- Improve core scheduling by fixing a bug in
sched_core_update_cookie() that caused unnecessary forced idling.
- Improve wakeup-balancing by allowing same-LLC wakeup of idle CPUs
for newly woken tasks.
- Fix a newidle balancing bug that introduced unnecessary wakeup
latencies.
ABI improvements/fixes:
- Do not check capabilities and do not issue capability check denial
messages when a scheduler syscall doesn't require privileges. (Such
as increasing niceness.)
- Add forced-idle accounting to cgroups too.
- Fix/improve the RSEQ ABI to not just silently accept unknown flags.
(No existing tooling is known to have learned to rely on the
previous behavior.)
- Depreciate the (unused) RSEQ_CS_FLAG_NO_RESTART_ON_* flags.
Optimizations:
- Optimize & simplify leaf_cfs_rq_list()
- Micro-optimize set_nr_{and_not,if}_polling() via try_cmpxchg().
Misc fixes & cleanups:
- Fix the RSEQ self-tests on RISC-V and Glibc 2.35 systems.
- Fix a full-NOHZ bug that can in some cases result in the tick not
being re-enabled when the last SCHED_RT task is gone from a
runqueue but there's still SCHED_OTHER tasks around.
- Various PREEMPT_RT related fixes.
- Misc cleanups & smaller fixes"
* tag 'sched-core-2022-08-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits)
rseq: Kill process when unknown flags are encountered in ABI structures
rseq: Deprecate RSEQ_CS_FLAG_NO_RESTART_ON_* flags
sched/core: Fix the bug that task won't enqueue into core tree when update cookie
nohz/full, sched/rt: Fix missed tick-reenabling bug in dequeue_task_rt()
sched/core: Always flush pending blk_plug
sched/fair: fix case with reduced capacity CPU
sched/core: Use try_cmpxchg in set_nr_{and_not,if}_polling
sched/core: add forced idle accounting for cgroups
sched/fair: Remove the energy margin in feec()
sched/fair: Remove task_util from effective utilization in feec()
sched/fair: Use the same cpumask per-PD throughout find_energy_efficient_cpu()
sched/fair: Rename select_idle_mask to select_rq_mask
sched, drivers: Remove max param from effective_cpu_util()/sched_cpu_util()
sched/fair: Decay task PELT values during wakeup migration
sched/fair: Provide u64 read for 32-bits arch helper
sched/fair: Introduce SIS_UTIL to search idle CPU based on sum of util_avg
sched: only perform capability check on privileged operation
sched: Remove unused function group_first_cpu()
sched/fair: Remove redundant word " *"
selftests/rseq: check if libc rseq support is registered
...
The cgroup_update_dfl_csses() function updates css associations when a
cgroup's subtree_control file is modified. Any changes made to a cgroup's
subtree_control file, however, will only affect its descendants but not
the cgroup itself. So there is no point in migrating csses associated
with that cgroup. We can skip them instead.
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
30312730bd ("cgroup: Add "no" prefixed mount options") added "no" prefixed
mount options to allow turning them off and 6a010a49b6 ("cgroup: Make
!percpu threadgroup_rwsem operations optional") added one more "no" prefixed
mount option. However, Michal pointed out that the "no" prefixed options
aren't necessary in allowing mount options to be turned off:
# grep group /proc/mounts
cgroup2 /sys/fs/cgroup cgroup2 rw,nosuid,nodev,relatime,nsdelegate,memory_recursiveprot 0 0
# mount -o remount,nsdelegate,memory_recursiveprot none /sys/fs/cgroup
# grep cgroup /proc/mounts
cgroup2 /sys/fs/cgroup cgroup2 rw,relatime,nsdelegate,memory_recursiveprot 0 0
Note that this is different from the remount behavior when the mount(1) is
invoked without the device argument - "none":
# grep cgroup /proc/mounts
cgroup2 /sys/fs/cgroup cgroup2 rw,nosuid,nodev,noexec,relatime,nsdelegate,memory_recursiveprot 0 0
# mount -o remount,nsdelegate,memory_recursiveprot /sys/fs/cgroup
# grep cgroup /proc/mounts
cgroup2 /sys/fs/cgroup cgroup2 rw,nosuid,nodev,noexec,relatime,nsdelegate,memory_recursiveprot 0 0
While a bit confusing, given that there is a way to turn off the options,
there's no reason to have the explicit "no" prefixed options. Let's remove
them.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
3942a9bd7b ("locking, rcu, cgroup: Avoid synchronize_sched() in
__cgroup_procs_write()") disabled percpu operations on threadgroup_rwsem
because the impiled synchronize_rcu() on write locking was pushing up the
latencies too much for android which constantly moves processes between
cgroups.
This makes the hotter paths - fork and exit - slower as they're always
forced into the slow path. There is no reason to force this on everyone
especially given that more common static usage pattern can now completely
avoid write-locking the rwsem. Write-locking is elided when turning on and
off controllers on empty sub-trees and CLONE_INTO_CGROUP enables seeding a
cgroup without grabbing the rwsem.
Restore the default percpu operations and introduce the mount option
"favordynmods" and config option CGROUP_FAVOR_DYNMODS for users who need
lower latencies for the dynamic operations.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Michal Koutn� <mkoutny@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Dmitry Shmidt <dimitrysh@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
We allow modifying these mount options via remount. Let's add "no" prefixed
variants so that they can be turned off too.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
cgroup_update_dfl_csses() write-lock the threadgroup_rwsem as updating the
csses can trigger process migrations. However, if the subtree doesn't
contain any tasks, there aren't gonna be any cgroup migrations. This
condition can be trivially detected by testing whether
mgctx.preloaded_src_csets is empty. Elide write-locking threadgroup_rwsem if
the subtree is empty.
After this optimization, the usage pattern of creating a cgroup, enabling
the necessary controllers, and then seeding it with CLONE_INTO_CGROUP and
then removing the cgroup after it becomes empty doesn't need to write-lock
threadgroup_rwsem at all.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Michal Koutný <mkoutny@suse.com>
4feee7d126 previously added per-task forced idle accounting. This patch
extends this to also include cgroups.
rstat is used for cgroup accounting, except for the root, which uses
kcpustat in order to bypass the need for doing an rstat flush when
reading root stats.
Only cgroup v2 is supported. Similar to the task accounting, the cgroup
accounting requires that schedstats is enabled.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lkml.kernel.org/r/20220629211426.3329954-1-joshdon@google.com
We have:
int cgroup_migrate_vet_dst(struct cgroup *dst_cgrp)
{
...
/* mixables don't care */
if (cgroup_is_mixable(dst_cgrp))
return 0;
/*
* If @dst_cgrp is already or can become a thread root or is
* threaded, it doesn't matter.
*/
if (cgroup_can_be_thread_root(dst_cgrp) || cgroup_is_threaded(dst_cgrp))
return 0;
...
}
but in fact the entry of cgroup_can_be_thread_root() covers case that
checking cgroup_is_mixable() as following:
static bool cgroup_can_be_thread_root(struct cgroup *cgrp)
{
/* mixables don't care */
if (cgroup_is_mixable(cgrp))
return true;
...
}
so explicitly checking in cgroup_migrate_vet_dst is unnecessary.
Signed-off-by: Lin Feng <linf@wangsu.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Each cset (css_set) is pinned by its tasks. When we're moving tasks around
across csets for a migration, we need to hold the source and destination
csets to ensure that they don't go away while we're moving tasks about. This
is done by linking cset->mg_preload_node on either the
mgctx->preloaded_src_csets or mgctx->preloaded_dst_csets list. Using the
same cset->mg_preload_node for both the src and dst lists was deemed okay as
a cset can't be both the source and destination at the same time.
Unfortunately, this overloading becomes problematic when multiple tasks are
involved in a migration and some of them are identity noop migrations while
others are actually moving across cgroups. For example, this can happen with
the following sequence on cgroup1:
#1> mkdir -p /sys/fs/cgroup/misc/a/b
#2> echo $$ > /sys/fs/cgroup/misc/a/cgroup.procs
#3> RUN_A_COMMAND_WHICH_CREATES_MULTIPLE_THREADS &
#4> PID=$!
#5> echo $PID > /sys/fs/cgroup/misc/a/b/tasks
#6> echo $PID > /sys/fs/cgroup/misc/a/cgroup.procs
the process including the group leader back into a. In this final migration,
non-leader threads would be doing identity migration while the group leader
is doing an actual one.
After #3, let's say the whole process was in cset A, and that after #4, the
leader moves to cset B. Then, during #6, the following happens:
1. cgroup_migrate_add_src() is called on B for the leader.
2. cgroup_migrate_add_src() is called on A for the other threads.
3. cgroup_migrate_prepare_dst() is called. It scans the src list.
4. It notices that B wants to migrate to A, so it tries to A to the dst
list but realizes that its ->mg_preload_node is already busy.
5. and then it notices A wants to migrate to A as it's an identity
migration, it culls it by list_del_init()'ing its ->mg_preload_node and
putting references accordingly.
6. The rest of migration takes place with B on the src list but nothing on
the dst list.
This means that A isn't held while migration is in progress. If all tasks
leave A before the migration finishes and the incoming task pins it, the
cset will be destroyed leading to use-after-free.
This is caused by overloading cset->mg_preload_node for both src and dst
preload lists. We wanted to exclude the cset from the src list but ended up
inadvertently excluding it from the dst list too.
This patch fixes the issue by separating out cset->mg_preload_node into
->mg_src_preload_node and ->mg_dst_preload_node, so that the src and dst
preloadings don't interfere with each other.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Mukesh Ojha <quic_mojha@quicinc.com>
Reported-by: shisiyuan <shisiyuan19870131@gmail.com>
Link: http://lkml.kernel.org/r/1654187688-27411-1-git-send-email-shisiyuan@xiaomi.com
Link: https://www.spinics.net/lists/cgroups/msg33313.html
Fixes: f817de9851 ("cgroup: prepare migration path for unified hierarchy")
Cc: stable@vger.kernel.org # v3.16+
Memory about struct psi_group is allocated by default for
each cgroup even if psi_disabled is true, in this case, these
allocated memory is waste, so alloc memory for struct psi_group
only when psi_disabled is false.
Signed-off-by: Chen Wandun <chenwandun@huawei.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Pull cgroup updates from Tejun Heo:
"Nothing too interesting. This adds cpu controller selftests and there
are a couple code cleanup patches"
* 'for-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup: remove the superfluous judgment
cgroup: Make cgroup_debug static
kseltest/cgroup: Make test_stress.sh work if run interactively
kselftest/cgroup: fix test_stress.sh to use OUTPUT dir
cgroup: Add config file to cgroup selftest suite
cgroup: Add test_cpucg_max_nested() testcase
cgroup: Add test_cpucg_max() testcase
cgroup: Add test_cpucg_nested_weight_underprovisioned() testcase
cgroup: Adding test_cpucg_nested_weight_overprovisioned() testcase
cgroup: Add test_cpucg_weight_underprovisioned() testcase
cgroup: Add test_cpucg_weight_overprovisioned() testcase
cgroup: Add test_cpucg_stats() testcase to cgroup cpu selftests
cgroup: Add new test_cpu.c test suite in cgroup selftests
Remove the superfluous judgment since the function is
never called for a root cgroup, as suggested by Tejun.
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Shida Zhang <zhangshida@kylinos.cn>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Make cgroup_debug static since it's only used in cgroup.c
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
There are 3 places where the cpu and node masks of the top cpuset can
be initialized in the order they are executed:
1) start_kernel -> cpuset_init()
2) start_kernel -> cgroup_init() -> cpuset_bind()
3) kernel_init_freeable() -> do_basic_setup() -> cpuset_init_smp()
The first cpuset_init() call just sets all the bits in the masks.
The second cpuset_bind() call sets cpus_allowed and mems_allowed to the
default v2 values. The third cpuset_init_smp() call sets them back to
v1 values.
For systems with cgroup v2 setup, cpuset_bind() is called once. As a
result, cpu and memory node hot add may fail to update the cpu and node
masks of the top cpuset to include the newly added cpu or node in a
cgroup v2 environment.
For systems with cgroup v1 setup, cpuset_bind() is called again by
rebind_subsystem() when the v1 cpuset filesystem is mounted as shown
in the dmesg log below with an instrumented kernel.
[ 2.609781] cpuset_bind() called - v2 = 1
[ 3.079473] cpuset_init_smp() called
[ 7.103710] cpuset_bind() called - v2 = 0
smp_init() is called after the first two init functions. So we don't
have a complete list of active cpus and memory nodes until later in
cpuset_init_smp() which is the right time to set up effective_cpus
and effective_mems.
To fix this cgroup v2 mask setup problem, the potentially incorrect
cpus_allowed & mems_allowed setting in cpuset_init_smp() are removed.
For cgroup v2 systems, the initial cpuset_bind() call will set the masks
correctly. For cgroup v1 systems, the second call to cpuset_bind()
will do the right setup.
cc: stable@vger.kernel.org
Signed-off-by: Waiman Long <longman@redhat.com>
Tested-by: Feng Tang <feng.tang@intel.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Pull driver core updates from Greg KH:
"Here is the set of driver core changes for 5.18-rc1.
Not much here, primarily it was a bunch of cleanups and small updates:
- kobj_type cleanups for default_groups
- documentation updates
- firmware loader minor changes
- component common helper added and take advantage of it in many
drivers (the largest part of this pull request).
All of these have been in linux-next for a while with no reported
problems"
* tag 'driver-core-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (54 commits)
Documentation: update stable review cycle documentation
drivers/base/dd.c : Remove the initial value of the global variable
Documentation: update stable tree link
Documentation: add link to stable release candidate tree
devres: fix typos in comments
Documentation: add note block surrounding security patch note
samples/kobject: Use sysfs_emit instead of sprintf
base: soc: Make soc_device_match() simpler and easier to read
driver core: dd: fix return value of __setup handler
driver core: Refactor sysfs and drv/bus remove hooks
driver core: Refactor multiple copies of device cleanup
scripts: get_abi.pl: Fix typo in help message
kernfs: fix typos in comments
kernfs: remove unneeded #if 0 guard
ALSA: hda/realtek: Make use of the helper component_compare_dev_name
video: omapfb: dss: Make use of the helper component_compare_dev
power: supply: ab8500: Make use of the helper component_compare_dev
ASoC: codecs: wcd938x: Make use of the helper component_compare/release_of
iommu/mediatek: Make use of the helper component_compare/release_of
drm: of: Make use of the helper component_release_of
...
Merge more updates from Andrew Morton:
"Various misc subsystems, before getting into the post-linux-next
material.
41 patches.
Subsystems affected by this patch series: procfs, misc, core-kernel,
lib, checkpatch, init, pipe, minix, fat, cgroups, kexec, kdump,
taskstats, panic, kcov, resource, and ubsan"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (41 commits)
Revert "ubsan, kcsan: Don't combine sanitizer with kcov on clang"
kernel/resource: fix kfree() of bootmem memory again
kcov: properly handle subsequent mmap calls
kcov: split ioctl handling into locked and unlocked parts
panic: move panic_print before kmsg dumpers
panic: add option to dump all CPUs backtraces in panic_print
docs: sysctl/kernel: add missing bit to panic_print
taskstats: remove unneeded dead assignment
kasan: no need to unset panic_on_warn in end_report()
ubsan: no need to unset panic_on_warn in ubsan_epilogue()
panic: unset panic_on_warn inside panic()
docs: kdump: add scp example to write out the dump file
docs: kdump: update description about sysfs file system support
arm64: mm: use IS_ENABLED(CONFIG_KEXEC_CORE) instead of #ifdef
x86/setup: use IS_ENABLED(CONFIG_KEXEC_CORE) instead of #ifdef
riscv: mm: init: use IS_ENABLED(CONFIG_KEXEC_CORE) instead of #ifdef
kexec: make crashk_res, crashk_low_res and crash_notes symbols always visible
cgroup: use irqsave in cgroup_rstat_flush_locked().
fat: use pointer to simple type in put_user()
minix: fix bug when opening a file with O_DIRECT
...
