Re: [Qemu-devel] [PATCH v2 for-3.0] tests/libqtest: Improve kill_qemu()

[Markus Armbruster <armbru@redhat.com>]

Eric Blake <eblake@redhat.com> writes:

> On 07/25/2018 11:17 AM, Markus Armbruster wrote:
>
>>>
>>> the output was produced by bash, which uses waitpid() - and therefore
>>> the fact that bash reports the core dump even when no core file is
>>> created is promising.
>>
>> Proof beats plausibility argument:
>>
>> $ cat wcordump.c
>
>> $ gcc -Wall -g -O wcordump.c
>> $ (ulimit -c unlimited; ./a.out)
>> sig 6 128
>> $ (ulimit -c 0; ./a.out)
>> sig 6 0
>
> Doesn't match my results:
>
> $ (ulimit -c 0; ./a.out)
> sig 6 128
>
> So, what's different between our two environments?
>
> kernel 4.17.7-100.fc27.x86_64

4.17.7-200.fc28.x86_64

> $ echo /proc/sys/kernel/core_*
> /proc/sys/kernel/core_pattern /proc/sys/kernel/core_pipe_limit
> /proc/sys/kernel/core_uses_pid
> $ cat /proc/sys/kernel/core_*
> |/usr/lib/systemd/systemd-coredump %P %u %g %s %t %c %e

This is "core" on my development boxes (I'm a happy caveman).

# cat /etc/sysctl.d/50-coredump.conf 
kernel.core_pattern=core

> 0
> 1
>
>>
>> Looks like WCOREDUMP() does depend on my ulimit -c.
>
> Or worse, that its behavior is kernel/environment-sensitive.

Looks like it.