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:

  e1b77c92981a5222 ("sched/kasan: remove stale KASAN poison after hotplug")

... by removing any stale KASAN stack poison immediately prior to
onlining a CPU.

Subsequently in commit:

  f1a0a376ca0c4ef1 ("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:

  63acd42c0d4942f7 ("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: f1a0a376ca0c4ef1 ("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/

CPU hotplug callbacks can fail and cause a rollback to the previous
state. These failures are silent and therefore hard to debug.

Add pr_debug() to the up and down paths which provide information about the
error code, the CPU and the failed state. The debug printks can be enabled
via kernel command line or sysfs.

[ tglx: Adopt to current mainline, massage printk and changelog ]

Signed-off-by: Dongli Zhang <dongli.zhang@oracle.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Qais Yousef <qais.yousef@arm.com>
Link: https://lore.kernel.org/r/20210409055316.1709-1-dongli.zhang@oracle.com

Use DEVICE_ATTR_*() helper instead of plain DEVICE_ATTR,
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210527141105.2312-1-yuehaibing@huawei.com

kernel/cpu.c:57: warning: cannot understand function prototype: 'struct cpuhp_cpu_state '
kernel/cpu.c:115: warning: cannot understand function prototype: 'struct cpuhp_step '
kernel/cpu.c:146: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * cpuhp_invoke_callback _ Invoke the callbacks for a given state
kernel/cpu.c:75: warning: Function parameter or member 'fail' not described in 'cpuhp_cpu_state'
kernel/cpu.c:75: warning: Function parameter or member 'cpu' not described in 'cpuhp_cpu_state'
kernel/cpu.c:75: warning: Function parameter or member 'node' not described in 'cpuhp_cpu_state'
kernel/cpu.c:75: warning: Function parameter or member 'last' not described in 'cpuhp_cpu_state'
kernel/cpu.c:130: warning: Function parameter or member 'list' not described in 'cpuhp_step'
kernel/cpu.c:130: warning: Function parameter or member 'multi_instance' not described in 'cpuhp_step'
kernel/cpu.c:158: warning: No description found for return value of 'cpuhp_invoke_callback'
kernel/cpu.c:1188: warning: No description found for return value of 'cpu_device_down'
kernel/cpu.c:1400: warning: No description found for return value of 'cpu_device_up'
kernel/cpu.c:1425: warning: No description found for return value of 'bringup_hibernate_cpu'

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210809223825.24512-1-rdunlap@infradead.org

Fixes the following W=1 kernel build warning(s):

 kernel/cpu.c:1949: warning: Function parameter or member 
  'name' not described in '__cpuhp_setup_state_cpuslocked'

Signed-off-by: Baokun Li <libaokun1@huawei.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210605063003.681049-1-libaokun1@huawei.com

Pull CPU hotplug fix from Thomas Gleixner:
 "A fix for the CPU hotplug and cpusets interaction:

  cpusets delegate the hotplug work to a workqueue to prevent a lock
  order inversion vs. the CPU hotplug lock. The work is not flushed
  before the hotplug operation returns which creates user visible
  inconsistent state. Prevent this by flushing the work after dropping
  CPU hotplug lock and before releasing the outer mutex which serializes
  the CPU hotplug related sysfs interface operations"

* tag 'smp-urgent-2021-06-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  cpu/hotplug: Cure the cpusets trainwreck

Alexey and Joshua tried to solve a cpusets related hotplug problem which is
user space visible and results in unexpected behaviour for some time after
a CPU has been plugged in and the corresponding uevent was delivered.

cpusets delegate the hotplug work (rebuilding cpumasks etc.) to a
workqueue. This is done because the cpusets code has already a lock
nesting of cgroups_mutex -> cpu_hotplug_lock. A synchronous callback or
waiting for the work to finish with cpu_hotplug_lock held can and will
deadlock because that results in the reverse lock order.

As a consequence the uevent can be delivered before cpusets have consistent
state which means that a user space invocation of sched_setaffinity() to
move a task to the plugged CPU fails up to the point where the scheduled
work has been processed.

The same is true for CPU unplug, but that does not create user observable
failure (yet).

It's still inconsistent to claim that an operation is finished before it
actually is and that's the real issue at hand. uevents just make it
reliably observable.

Obviously the problem should be fixed in cpusets/cgroups, but untangling
that is pretty much impossible because according to the changelog of the
commit which introduced this 8 years ago:

 3a5a6d0c2b03("cpuset: don't nest cgroup_mutex inside get_online_cpus()")

the lock order cgroups_mutex -> cpu_hotplug_lock is a design decision and
the whole code is built around that.

So bite the bullet and invoke the relevant cpuset function, which waits for
the work to finish, in _cpu_up/down() after dropping cpu_hotplug_lock and
only when tasks are not frozen by suspend/hibernate because that would
obviously wait forever.

Waiting there with cpu_add_remove_lock, which is protecting the present
and possible CPU maps, held is not a problem at all because neither work
queues nor cpusets/cgroups have any lockchains related to that lock.

Waiting in the hotplug machinery is not problematic either because there
are already state callbacks which wait for hardware queues to drain. It
makes the operations slightly slower, but hotplug is slow anyway.

This ensures that state is consistent before returning from a hotplug
up/down operation. It's still inconsistent during the operation, but that's
a different story.

Add a large comment which explains why this is done and why this is not a
dump ground for the hack of the day to work around half thought out locking
schemes. Document also the implications vs. hotplug operations and
serialization or the lack of it.

Thanks to Alexy and Joshua for analyzing why this temporary
sched_setaffinity() failure happened.

Fixes: 3a5a6d0c2b03("cpuset: don't nest cgroup_mutex inside get_online_cpus()")
Reported-by: Alexey Klimov <aklimov@redhat.com>
Reported-by: Joshua Baker <jobaker@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Alexey Klimov <aklimov@redhat.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/87tuowcnv3.ffs@nanos.tec.linutronix.de

It is unnecessary to invoke per_cpu_ptr() everytime to access cpuhp_state.
Use the available pointer instead.

Signed-off-by: Yuan ZhaoXiong <yuanzhaoxiong@baidu.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lore.kernel.org/r/1621776690-13264-1-git-send-email-yuanzhaoxiong@baidu.com

Vincent reported that for states with a NULL startup/teardown function
we do not call cpuhp_invoke_callback() (because there is none) and as
such we'll not update the cpu_dying() state.

The stale cpu_dying() can eventually lead to triggering BUG().

Rectify this by updating cpu_dying() in the exact same places the
hotplug machinery tracks its directional state, namely
cpuhp_set_state() and cpuhp_reset_state().

Reported-by: Vincent Donnefort <vincent.donnefort@arm.com>
Suggested-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Donnefort <vincent.donnefort@arm.com>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lkml.kernel.org/r/YH7r+AoQEReSvxBI@hirez.programming.kicks-ass.net

Introduce a cpumask that indicates (for each CPU) what direction the
CPU hotplug is currently going. Notably, it tracks rollbacks. Eg. when
an up fails and we do a roll-back down, it will accurately reflect the
direction.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lkml.kernel.org/r/20210310150109.151441252@infradead.org