Re: [PATCH v3 2/2] cpuhp: Expedite RCU grace periods during SMT operations

[Vishal Chourasia <vishalc@linux.ibm.com>]

Hi Paul, Thank you for your response. 
Sorry, I could not revert back quicker.

As I wanted to understand what's happening behind the scenes after the
cpuhp kthread blocks upon execution of synchronize_rcu(). So I did a
little more digging.

On a 320 CPU system, (SMT8 to SMT4) operation takes >1 minute to complete.
160 CPUs are offlined one by one. 

In total 321 synchronize_rcu() calls are invoked taking ~125ms to finish
(ftrace option sleep-time set).

3298110.851011 |   316)  cpuhp/3-1614  |               |  synchronize_rcu() {
3298111.010125 |   316)  cpuhp/3-1614  | @ 159112.9 us |  }
--
3298111.020432 |     0) kworker-29406  |               |  synchronize_rcu() {
3298111.190132 |     0) kworker-29406  | @ 169699.4 us |  }
--
3298111.191327 |   317)  cpuhp/3-1619  |               |  synchronize_rcu() {
3298111.350129 |   317)  cpuhp/3-1619  | @ 158801.9 us |  }
--
3298111.360263 |     0) kworker-29406  |               |  synchronize_rcu() {
3298111.530137 |     0) kworker-29406  | @ 169874.5 us |  }
--
3298111.531098 |   318)  cpuhp/3-1624  |               |  synchronize_rcu() {
3298111.650128 |   318)  cpuhp/3-1624  | @ 119029.8 us |  }

Breakdown of the time spent during a single synchronize_rcu() during the
invocation of sched_cpu_deactivate callback (CPU 4 was offlined)

Summary:
--> cpuhp_enter (sched_cpu_deactivate)
CB registration → AccWaitCB        ~10ms     Waiting for softirq tick on CPU 4
GP 220685125: FQS scan 1           ~10ms     Tick delay + scan (all clear except CPU 260, 
                                                        rcu_gp_kthread is running on CPU 260)
GP 220685125: wait for CPU 260     ~30msi    FQS sleep interval, CPU 260 not yet reported
GP 220685125: FQS scan 2 + end     ~0.02ms   CPU 260 clears
GP 220685129: FQS scan 1           ~30ms     Tick delay + full scan (same: CPU 260 holdout)
GP 220685129: wait for CPU 260     ~30ms     Same pattern
GP 220685129: FQS scan 2 + end     ~0.02ms   CPU 260 clears
CB invocation + wakeup             ~10ms     Softirq tick invokes wakeme_after_rcu
destroy_sched_domains_rcu queueing ~8ms322   call_rcu() callbacks
<-- cpuhp_exit (sched_cpu_deactivate)

I have collected some rcu static tracepoint data, which I am currently
going through.

On Fri, Mar 06, 2026 at 07:12:04AM -0800, Paul E. McKenney wrote:
> On Fri, Mar 06, 2026 at 11:14:13AM +0530, Vishal Chourasia wrote:
> > On Mon, Mar 02, 2026 at 05:17:16PM +0530, Samir M wrote:
> > > 
> > > On 27/02/26 6:43 am, Joel Fernandes wrote:
> > > > On Wed, Feb 18, 2026 at 02:09:18PM +0530, Vishal Chourasia wrote:
> > > > > Expedite synchronize_rcu during the SMT mode switch operation when
> > > > > initiated via /sys/devices/system/cpu/smt/control interface
> > > > >
> > > > After the locking related changes in patch 1, is expediting still required? I
> > Yes.
> > > > am just a bit concerned that we are papering over the real issue of over
> > > > usage of synchronize_rcu() (which IIRC we discussed in earlier versions of
> > > > the patches that reducing the number of lock acquire/release was supposed to
> > > > help.)
> > At present, I am not sure about the underlying issue. So far what I have
> > found is when synchronize_rcu() is invoked, it marks the start of a new
> > grace period number, say A. Thread invoking synchronize_rcu() blocks
> > until all CPUs have reported QS for GP "A". There is a rcu grace period
> > kthread that runs periodically looping over a CPU list to figure out all
> > CPUs have reported QS. In the trace, I find some CPUs reporting QS for
> > sequence number way back in the past for ex. A - N where N is > 10.
> 
> This can happen when a CPU goes idle for multiple grace periods, then
> wakes up in the middle of a later grace period.  This is (or at least is
> supposed to be) harmless because a quiescent state was reported on that
> CPU's behalf when RCU noticed that it was idle.  The report is quashed
If it is harmless, can we consider just expediting the smt mode switch
operation via smt/control file [1].

Thanks, vishalc

[1] https://lore.kernel.org/all/20260218083915.660252-6-vishalc@linux.ibm.com/

> when RCU notices that the quiescent state being reported is for a grace
> period that has already completed.  Grace-period counter wrap is handled
> by the infamous ->gpwrap field in the rcu_data structure.

> 
> I have seen N having four digits, with deep embedded devices being most
> likely to have extremely large values of N.
> 
> 							Thanx, Paul
> 
> > > > Could you provide more justification of why expediting these sections is
> > > > required if the locking concerns were addressed? It would be great if you can
> > > > provide performance numbers with only the first patch and without the second
> > > > patch. That way we can quantify this patch.
> > > > 
> > > > 
> > > SMT Mode    | Without Patch(Base) | both patch applied | % Improvement  |
> > > ------------------------------------------------------------------------|
> > > SMT=off     | 16m 13.956s         |     6m 18.435s     |  +61.14 %      |
> > > SMT=on      | 12m 0.982s          |     5m 59.576s     |  +50.10 %      |
> > > 
> > > When I tested the below patch independently, I did not observe any
> > > improvements for either smt=on or smt=off. However, in the smt=off scenario,
> > > I encountered hung task splats (with call traces), where some threads were
> > > blocked on cpus_read_lock. Please also refer to the attached call trace
> > > below.
> > > Patch 1:
> > > https://lore.kernel.org/all/20260218083915.660252-4-vishalc@linux.ibm.com/
> > > 
> > > SMT Mode    | Without Patch(Base) | just patch 1 applied   | % Improvement 
> > > |
> > > ----------------------------------------------------------------------------|
> > > SMT=off     | 16m 13.956s         |     16m 9.793s         |  +0.43 %     
> > >  |
> > > SMT=on      | 12m 0.982s          |     12m 19.494s        |  -2.57 %     
> > >  |
> > > 
> > > 
> > > Call traces:
> > > 12377] [  T8746]    Tainted: G      E 7.0.0-rc1-150700.51-default-dirty #1
> > > [ 1477.612384] [  T8746] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
> > > disables this message.
> > > [ 1477.612389] [  T8746] task:systemd     state:D stack:0   pid:1  tgid:1 
> > >  ppid:0   task_flags:0x400100 flags:0x00040000
> > > [ 1477.612397] [  T8746] Call Trace:
> > > [ 1477.612399] [  T8746] [c00000000cc0f4f0] [0000000000100000] 0x100000
> > > (unreliable)
> > > [ 1477.612416] [  T8746] [c00000000cc0f6a0] [c00000000001fe5c]
> > > __switch_to+0x1dc/0x290
> > > [ 1477.612425] [  T8746] [c00000000cc0f6f0] [c0000000012598ac]
> > > __schedule+0x40c/0x1a70
> > > [ 1477.612433] [  T8746] [c00000000cc0f840] [c00000000125af58]
> > > schedule+0x48/0x1a0
> > > [ 1477.612439] [  T8746] [c00000000cc0f870] [c0000000002e27b8]
> > > percpu_rwsem_wait+0x198/0x200
> > > [ 1477.612445] [  T8746] [c00000000cc0f8f0] [c000000001262930]
> > > __percpu_down_read+0xb0/0x210
> > > [ 1477.612449] [  T8746] [c00000000cc0f930] [c00000000022f400]
> > > cpus_read_lock+0xc0/0xd0
> > > [ 1477.612456] [  T8746] [c00000000cc0f950] [c0000000003a6398]
> > > cgroup_procs_write_start+0x328/0x410
> > > [ 1477.612462] [  T8746] [c00000000cc0fa00] [c0000000003a9620]
> > > __cgroup_procs_write+0x70/0x2c0
> > > [ 1477.612468] [  T8746] [c00000000cc0fac0] [c0000000003a98e8]
> > > cgroup_procs_write+0x28/0x50
> > > [ 1477.612473] [  T8746] [c00000000cc0faf0] [c0000000003a1624]
> > > cgroup_file_write+0xb4/0x240
> > > [ 1477.612478] [  T8746] [c00000000cc0fb50] [c000000000853ba8]
> > > kernfs_fop_write_iter+0x1a8/0x2a0
> > > [ 1477.612485] [  T8746] [c00000000cc0fba0] [c000000000733d5c]
> > > vfs_write+0x27c/0x540
> > > [ 1477.612491] [  T8746] [c00000000cc0fc50] [c000000000734350]
> > > ksys_write+0x80/0x150
> > > [ 1477.612495] [  T8746] [c00000000cc0fca0] [c000000000032898]
> > > system_call_exception+0x148/0x320
> > > [ 1477.612500] [  T8746] [c00000000cc0fe50] [c00000000000d6a0]
> > > system_call_common+0x160/0x2c4
> > > [ 1477.612506] [  T8746] ---- interrupt: c00 at 0x7fffa8f73df4
> > > [ 1477.612509] [  T8746] NIP: 00007fffa8f73df4 LR: 00007fffa8eb6144 CTR:
> > > 0000000000000000
> > > [ 1477.612512] [  T8746] REGS: c00000000cc0fe80 TRAP: 0c00 Tainted: G     
> > > E    (7.0.0-rc1-150700.51-default-dirty)
> > > [ 1477.612515] [  T8746] MSR: 800000000000d033 <SF,EE,PR,ME,IR,DR,RI,LE> CR:
> > > 28002288 XER: 00000000
> > > 
> > > 
> > 
> > Default timeout is set to 8 mins.
> > 
> > $ grep . /proc/sys/kernel/hung_task_timeout_secs
> > /proc/sys/kernel/hung_task_timeout_secs:480
> > 
> > Now that cpus_write_lock is taken once, and SMT mode switch can take
> > tens of minutes to complete and relinquish the lock, threads waiting on 
> > cpus_read_lock will be blocked for this entire duration.
> > 
> > Although there were no splats observed for "both patch applied" case
> > the issue still remains.
> > 
> > regards,
> > vishal