Re: [PATCH] powerpc/stacktrace: fix raise_backtrace_ipi() logic

[Nathan Lynch <nathanl@linux.ibm.com>]

Michael Ellerman <mpe@ellerman.id.au> writes:
> Nathan Lynch <nathanl@linux.ibm.com> writes:
>> Hi Michael,
>>
>> Michael Ellerman <mpe@ellerman.id.au> writes:
>>> Nathan Lynch <nathanl@linux.ibm.com> writes:
>>>> When smp_send_safe_nmi_ipi() indicates that the target CPU has
>>>> responded to the IPI, skip the remote paca inspection
>>>> fallback. Otherwise both the sending and target CPUs attempt the
>>>> backtrace, usually creating a misleading ("didn't respond to backtrace
>>>> IPI" is wrong) and interleaved mess:
>>>
>>> Thanks for fixing my bugs for me :)
>>>
>>
>> Thanks for your review! I was beginning to think I had missed some
>> subtletly here, thanks for illustrating it.
>>
>> I'll run with your proposed change below for the problem I'm working.
>
> Thanks. I did test it a bit with the test_lockup module, but some real
> world testing would be good too.

Been running with this to work some LPM issues and can confirm it waits
the intended amount of time before falling back to a remote stack walk,
avoiding interleaved traces from source and target CPUs. You can add my
tested-by, thanks.