RFR(XXS): 8224793: os::die() does not honor CreateCoredumpOnCrash option

[Daniel D. Daugherty]

On 5/31/19 8:10 PM, Kim Barrett wrote:
>> On May 31, 2019, at 4:25 PM, Daniel D. Daugherty <daniel.daugherty at oracle.com> wrote:
>>
>> Greetings,
>>
>> I've updated the fix based on the code review discussions with Kim and
>> David H. I think it's easier to review using the full webrev, but if
>> you're curious exactly how the code evolved, then there is also an
>> incremental webrev.
>>
>> Full webrev:
>> http://cr.openjdk.java.net/~dcubed/8224793-webrev/2-for-jdk-jdk13.full/
>>
>> Incremental webrev:
>> http://cr.openjdk.java.net/~dcubed/8224793-webrev/2-for-jdk-jdk13.inc/
>>
>> I have Mach5 Tier[1-3] testing running right now and it is looking good.
>>
>> Dan
> Change looks good.

Thanks.


> Though I'd really like to see these 4 copies of the same 11 lines of
> code merged into one definition in os_posix.  The extra comment
> from os_bsd could be brought along to explain the use of SIGKILL
> rather than _exit.

I think I'd rather that be done by another bug fix. I'm done with
this one. It was supposed to be the simple one of the two that I
was trying to push this week...


> BTW, the Windows implementation of os::die never dumps core.  It
> "just" calls _exit (after a large blob of code to work around a bug
> that I made the mistake of looking at).

That was my interpretation of the Win* code, but I didn't want to
add that platform to this simple fix either...

Dan