Re: [RFC PATCH] mce: prevent concurrent polling of MCE events

[Aristeu Rozanski <aris@ruivo.org>]

On Tue, Jan 17, 2023 at 05:42:36PM +0000, Luck, Tony wrote:
> >
> >               if (mce_intel_cmci_poll()) {
> >                       iv = mce_adjust_timer(iv);
> 
> If the CMCI polling interrupts on those large number of CPUs are
> staggered at different times, then this should be fine. But if a largish
> number of CPUs get in lockstep with each other, then this could get
> ugly.  I.e. 200 CPUs all take the "CMCI poll" interrupt together. Then
> get stuck in a convoy here as one at a time step through checking
> the machine check banks.

Yes, but could change the patch to include mce_available() into the
protection of the lock. It should cleared once machine_check_poll() clears
the bank state, no?

> One option to avoid this might be to change from a fixed five minute
> polling interval to "five minute plus/minus rand(50) jiffies". Then even
> if some CPUs did sync up, they'd quickly diverge again.

That'd be fine for me as well. Perhaps initialize the polling interval using
cpu numbers to space them then random for rescheduling?

-- 
Aristeu