RFR: 8181317: SA: Enable corefile debugging on Portola for gcore dumped files
mikael.vidstedt at oracle.com
Mon Jun 19 19:58:53 UTC 2017
Looks good. Thanks for using the LIBC define!
> On Jun 19, 2017, at 3:05 AM, Jini George <jini.george at oracle.com> wrote:
> Thanks much, Poonam and Mikael. I have a revised webrev here to cater to this comment. Using
> OPENJDK_TARGET_LIBC here.
> Please do take a look.
> Mikael, it makes sense to ask the Alpine guys about the /proc/<pid>/maps text segment size discrepancy. I plan on doing that.
> - Jini.
> On 6/15/2017 5:04 AM, Mikael Vidstedt wrote:
>>> On Jun 14, 2017, at 8:00 AM, Poonam Parhar <poonam.bajaj at oracle.com> wrote:
>>> Hello Jini,
>>> The code changes look good!
>>> I was wondering if we should change the name of the flag INCLUDE_SA_ATTACH. Its current usage in the code appears to be for excluding the code not working on Alpine, but its name indicates that it is to enable/disable the SA attach mechanism.
>> INCLUDE_SA_ATTACH is indeed supposed to control the inclusion of the attach (ps_proc) functionality, independent of if the JDK is built for musl or not. It does seem like the use introduced by this change is different from the others in that it’s actually doing something musl specific, so this code should use some other mechanism for excluding the code.
>> From the findings in the bug I’m also wondering if this is a bug we should let the Alpine folks know about?
>>>> -----Original Message-----
>>>> From: Jini George
>>>> Sent: Monday, June 12, 2017 7:25 AM
>>>> To: portola-dev at openjdk.java.net
>>>> Subject: RFR: 8181317: SA: Enable corefile debugging on Portola for
>>>> gcore dumped files
>>>> Requesting a review for a very small change to avoid an address
>>>> conflict check for enabling core file debugging on Portola for gcore
>>>> generated corefiles.
>>>> Webrev: http://cr.openjdk.java.net/~jgeorge/8181317/webrev.00/
>>>> The analysis has been detailed in the bug.
More information about the portola-dev