Re: [PATCH] mm,hwpoison: non-current task should be checked early_kill for force_early

["HORIGUCHI NAOYA(堀口 直也)" <naoya.horiguchi@nec.com>]

On Tue, Jan 19, 2021 at 02:04:18PM +0800, Aili Yao wrote:
> 
> > Thanks, so which interface did you use for error injection? I guess first
> > you used /sys/devices/system/memory/hard_offline_page, but if it's true,
> > then the error event should be action optional (no MF_ACTION_REQUIRED set).
> > So now I'm wondering why you are observing action required events?
> > My another guess is that you might have used mce-inject tool, if that's true,
> > please use hard_offline_page, then current kernel code should properly send
> > SIGBUS to dedicated process.
> > 
> The test has no relation to ce and hard_offline_page, sorry for misleading.
> 
> if the test code will compiled to my_test bin, here is my script:
> ./my_test hola salut servus test haaa     --- this case no early-kill flag set
> or
> ./my_test -s hola salut servus test haaa  --- this case early-kill is set.
> 
> there must be a process names "test" which will trigger the UE;
> 
> when the my_test start, the shared page's physical address will be printed;
> In another console, I will use einj module to inject the 0X10 LEVEL error to this
> physical address.

Ah, OK, so the problem is becoming clearer, thanks.  I'm feeling positive
to change code to fall back to find_early_kill_thread().
So I'll take a look on v2.

Thanks,
Naoya Horiguchi