x86/mce: Make mce_timed_out() identify holdout CPUs
authorPaul E. McKenney <paulmck@kernel.org>
Thu, 24 Dec 2020 01:04:19 +0000 (17:04 -0800)
committerBorislav Petkov <bp@suse.de>
Fri, 8 Jan 2021 17:00:09 +0000 (18:00 +0100)
commit7bb39313cd6239e7eb95198950a02b4ad2a08316
tree5deba9f86d5be8a799c5a842d1e15ce51def766d
parente71ba9452f0b5b2e8dc8aa5445198cd9214a6a62
x86/mce: Make mce_timed_out() identify holdout CPUs

The

  "Timeout: Not all CPUs entered broadcast exception handler"

message will appear from time to time given enough systems, but this
message does not identify which CPUs failed to enter the broadcast
exception handler. This information would be valuable if available,
for example, in order to correlate with other hardware-oriented error
messages.

Add a cpumask of CPUs which maintains which CPUs have entered this
handler, and print out which ones failed to enter in the event of a
timeout.

 [ bp: Massage. ]

Reported-by: Jonathan Lemon <bsd@fb.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20210106174102.GA23874@paulmck-ThinkPad-P72
arch/x86/kernel/cpu/mce/core.c