cpu/hotplug: wait for cpuset_hotplug_work to finish on cpu online
Subsystems (2)
RFC PATCH 1/1 cpu/hotplug: wait for cpuset_hotplug_work to finish on cpu online
When a CPU offlined and onlined via device_offline() and device_online() the userspace gets uevent notification. If, after receiving uevent, userspace executes sched_setaffinity() on some task trying to move it to a recently onlined CPU, then it will fail with -EINVAL. Userspace needs to wait around 5..30 ms before sched_setaffinity() will succeed for recently onlined CPU after receiving uevent.
If in_mask for sched_setaffinity() has only recently onlined CPU, it quickly fails with such flow:
sched_setaffinity()
cpuset_cpus_allowed()
guarantee_online_cpus() <-- cs->effective_cpus mask does not
contain recently onlined cpu
cpumask_and() <-- final new_mask is empty
__set_cpus_allowed_ptr()
cpumask_any_and_distribute() <-- returns dest_cpu equal to nr_cpu_ids returns -EINVAL
Cpusets are updated using workqueue from cpuset_update_active_cpus() which in its turn is called from cpu hotplug callback sched_cpu_activate() hence the delay observable by sched_setaffinity().
Out of line uevent can be avoided if we will ensure that cpuset_hotplug_work has run to completion using cpuset_wait_for_hotplug() after onlining the cpu in cpu_up(). Unfortunately, the execution time of
echo 1 > /sys/devices/system/cpu/cpuX/online roughly doubled with this change (on my test machine).
---
The commit "cpuset: Make cpuset hotplug synchronous" would also get rid of the early uevent but it was reverted.
The nature of this bug is also described here (with different consequences):
https://lore.kernel.org/lkml/[email protected]/
Reproducer: https://gitlab.com/0xeafffffe/xlam
It could be that I missed the correct place for cpuset synchronisation and it should be done in cpu_device_up() instead.
I also in doubts if we need cpuset_wait_for_hotplug() in cpuhp_online_cpu_device() since an online uevent is sent there too.
Currently with such change the reproducer code continues to work without issues. The idea is to avoid the situation when userspace receives the event about onlined CPU which is not ready to take tasks for a while after uevent.
kernel/cpu.c | 3 +++ 1 file changed, 3 insertions(+)
kernel/cpu.c +3 -0
@@ -15,6 +15,7 @@#include <linux/sched/smt.h> #include <linux/unistd.h> #include <linux/cpu.h> +#include <linux/cpuset.h> #include <linux/oom.h> #include <linux/rcupdate.h> #include <linux/export.h>
@@ -1275,6 +1276,8 @@ static int cpu_up(unsigned int cpu, enum cpuhp_state target) } err =_cpu_up (cpu,0 , target); +if (!err) +cpuset_wait_for_hotplug (); out:cpu_maps_update_done ();return err;
Signature · 1 line
2.26.2
Peter Zijlstra
maintainerQuoted 6 lines · Alexey Klimov
When a CPU offlined and onlined via
device_offline()anddevice_online()the userspace gets uevent notification. If, after receiving uevent, userspace executessched_setaffinity()on some task trying to move it to a recently onlined CPU, then it will fail with-EINVAL. Userspace needs to wait around 5..30 ms beforesched_setaffinity()will succeed for recently onlined CPU after receiving uevent.
Right.
Quoted 3 lines
Unfortunately, the execution time of
echo 1 >/sys/devices/system/cpu/cpuX/onlineroughly doubled with this change (on my test machine).
Nobody cares, it's hotplug, it's supposed to be slow :-) That is, we fundamentally shift the work _to_ the hotplug path, so as to keep everybody else fast.
Quoted 2 lines
The nature of this bug is also described here (with different consequences):
https://lore.kernel.org/lkml/[email protected]/
Yeah, pesky deadlocks.. someone was going to try again.
Quoted 16 lines
kernel/cpu.c | 3 +++ 1 file changed, 3 insertions(+)kernel/cpu.c +1
-0@@ -15,6 +15,7 @@#include <linux/sched/smt.h> #include <linux/unistd.h> #include <linux/cpu.h> +#include <linux/cpuset.h> #include <linux/oom.h> #include <linux/rcupdate.h> #include <linux/export.h>
kernel/cpu.c static int cpu_up(unsigned int cpu, enum cpuhp_state target) · line 1283
Quoted 8 lines
kernel/cpu.c +2
-0@@ -1275,6 +1276,8 @@ static int cpu_up(unsigned int cpu, enum cpuhp_state target) } err =_cpu_up (cpu,0 , target); +if (!err) +cpuset_wait_for_hotplug (); out:cpu_maps_update_done ();return err;
My only consideration is if doing that flush under cpu_add_remove_lock() is wise.
Quoted 4 lines · Peter Zijlstra
Quoted 2 lines
The nature of this bug is also described here (with different consequences):
https://lore.kernel.org/lkml/[email protected]/Yeah, pesky deadlocks.. someone was going to try again.
I dug up the synchronous patch
https://lore.kernel.org/lkml/1579878449-10164-1-git-send-email-prsood-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org/
but surprisingly wasn't able to reproduce the lockdep splat from
https://lore.kernel.org/lkml/F0388D99-84D7-453B-9B6B-EEFF0E7BE4CC-J5quhbR+WMc@public.gmane.org/
even though I could hit it a few weeks ago. I'm going to try to mess with it later, but don't let me hold this up.
Quoted 8 lines · Daniel Jordan
Quoted 4 lines · Peter Zijlstra
Quoted 2 lines
The nature of this bug is also described here (with different consequences):
https://lore.kernel.org/lkml/[email protected]/Yeah, pesky deadlocks.. someone was going to try again.
I dug up the synchronous patch
Quoted 7 lines
https://lore.kernel.org/lkml/1579878449-10164-1-git-send-email-prsood-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org/but surprisingly wasn't able to reproduce the lockdep splat from
https://lore.kernel.org/lkml/F0388D99-84D7-453B-9B6B-EEFF0E7BE4CC-J5quhbR+WMc@public.gmane.org/even though I could hit it a few weeks ago.
oh okay, you need to mount a legacy cpuset hierarchy.
So as the above splat shows, making cpuset_hotplug_workfn() synchronous means cpu_hotplug_lock (and "cpuhp_state-down") can be acquired before cgroup_mutex.
But there are at least four cgroup paths that take the locks in the opposite order. They're all the same, they take cgroup_mutex and then cpu_hotplug_lock later on to modify one or more static keys.
cpu_hotplug_lock should probably be ahead of cgroup_mutex because the latter is taken in a hotplug callback, and we should keep the static branches in cgroup, so the only way out I can think of is moving cpu_hotplug_lock to just before cgroup_mutex is taken and switching to _cpuslocked flavors of the static key calls.
lockdep quiets down with that change everywhere, but it puts another big lock around a lot of cgroup paths. Seems less heavyhanded to go with this RFC. What do you all think?
Absent further discussion, Alexey, do you plan to post another version?
Quoted 34 lines · Daniel Jordan
Quoted 8 lines · Daniel Jordan
Quoted 4 lines · Peter Zijlstra
Quoted 2 lines
The nature of this bug is also described here (with different consequences):
https://lore.kernel.org/lkml/[email protected]/Yeah, pesky deadlocks.. someone was going to try again.
I dug up the synchronous patch
Quoted 7 lines
https://lore.kernel.org/lkml/1579878449-10164-1-git-send-email-prsood-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org/but surprisingly wasn't able to reproduce the lockdep splat from
https://lore.kernel.org/lkml/F0388D99-84D7-453B-9B6B-EEFF0E7BE4CC-J5quhbR+WMc@public.gmane.org/even though I could hit it a few weeks ago.
oh okay, you need to mount a legacy cpuset hierarchy.
So as the above splat shows, making
cpuset_hotplug_workfn()synchronous meanscpu_hotplug_lock(and "cpuhp_state-down") can be acquired beforecgroup_mutex.But there are at least four cgroup paths that take the locks in the opposite order. They're all the same, they take
cgroup_mutexand thencpu_hotplug_locklater on to modify one or more static keys.
cpu_hotplug_lockshould probably be ahead ofcgroup_mutexbecause the latter is taken in a hotplug callback, and we should keep the static branches in cgroup, so the only way out I can think of is movingcpu_hotplug_lockto just beforecgroup_mutexis taken and switching to_cpuslockedflavors of the static key calls.
Quoted 3 lines
lockdep quiets down with that change everywhere, but it puts another big lock around a lot of cgroup paths. Seems less heavyhanded to go with this RFC. What do you all think?
Daniel, thank you for taking a look. I don't mind reviewing+testing another approach that you described.
Quoted 1 line
Absent further discussion, Alexey, do you plan to post another version?
I plan to update this patch and re-send in the next couple of days. It looks like it might be a series of two patches. Sorry for delays.
Best regards,
Alexey
Quoted 2 lines · Alexey Klimov
Daniel, thank you for taking a look. I don't mind reviewing+testing another approach that you described.
Eh, I like yours better :)
Quoted 4 lines
Quoted 1 line
Absent further discussion, Alexey, do you plan to post another version?
I plan to update this patch and re-send in the next couple of days. It looks like it might be a series of two patches. Sorry for delays.
Not at all, this is just something I'm messing with when I have the time.
Quoted 2 lines · Alexey Klimov
I also in doubts if we need
cpuset_wait_for_hotplug()incpuhp_online_cpu_device()since an online uevent is sent there too.
We do need it there if we go with this fix. Your reproducer hits the same issue when it's changed to exercise smt/control instead of cpuN/online.