Is JDK-8291919 refixed in JDK11u?

[shanghe chen]

OK, I get it now. Thank you for your detailed explanation!

陈上河
________________________________
发件人: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>
发送时间: Thursday, February 22, 2024 12:18:09 AM
收件人: shanghe chen <chenshanghe at outlook.com>; jdk-dev at openjdk.org <jdk-dev at openjdk.org>; jdk-updates-dev <jdk-updates-dev-retn at openjdk.org>
主题: RE: Is JDK-8291919 refixed in JDK11u?


Hi,

The issue JDK-8291919 you are referring to is caused by “JDK-8279219 [REDO] C2 crash when allocating array of size too large”. By backing out JDK-8279219, which was done by JDK-8292260, the excess memory consumption is fixed. The excess memory consumption issue was in 11 from 11.0.16 to 11.0.16.1.  As you have problems with 11.0.14.1, you cannot see this  bug.

Further, “JDK-8292301<https://bugs.openjdk.org/browse/JDK-8292301> [REDO v2] C2 crash when allocating array of size too large”, which is not backported to 11u as you found out, fixes a crash. So backporting JDK-8292301 will not fix your issue.

Best regards, Goetz.



From: jdk-dev <jdk-dev-retn at openjdk.org> On Behalf Of shanghe chen
Sent: Mittwoch, 21. Februar 2024 07:25
To: jdk-dev at openjdk.org
Subject: Is JDK-8291919 refixed in JDK11u?



Some people who received this message don't often get email from chenshanghe at outlook.com<mailto:chenshanghe at outlook.com>. Learn why this is important<https://aka.ms/LearnAboutSenderIdentification>

Hi! We occionally occurred in to docker oom with a java service running inside docker with java version



openjdk version "11.0.14.1" 2022-02-08

OpenJDK Runtime Environment Temurin-11.0.14.1+1 (build 11.0.14.1+1)

OpenJDK 64-Bit Server VM Temurin-11.0.14.1+1 (build 11.0.14.1+1, mixed mode)



Inspecting dmesg log we found the following lines:



[Mon Feb 19 13:14:16 2024] IPv6: ADDRCONF(NETDEV_CHANGE): lxca2a678a797cd: link becomes ready

[Mon Feb 19 13:38:25 2024] C2 CompilerThre invoked oom-killer: gfp_mask=0x6000c0(GFP_KERNEL), nodemask=(null), order=0, oom_score_adj=-998

[Mon Feb 19 13:38:25 2024] C2 CompilerThre cpuset=2cc0b0ba8e280972d0d5269810c52d911a5227aa5d0b4dbc4da6b7003988000d mems_allowed=0

[Mon Feb 19 13:38:25 2024] CPU: 12 PID: 51367 Comm: C2 CompilerThre Tainted: G        W  O      4.19.118-2.el7.centos.x86_64 #1



And after some googling we found

- https://mail.openjdk.org/pipermail/hotspot-runtime-dev/2019-March/033112.html

- and https://bugs.openjdk.org/browse/JDK-8291919

- and https://bugs.openjdk.org/browse/JDK-8292260

- and https://bugs.openjdk.org/browse/JDK-8292301



JDK-8292260 says that the fix in JDK-8291919 has beed reverted. And then JDK-8292301 fixed this again.

But I cannot find any commits related to JDK-8292301 in jdk11u. Has it been fixed in jdk11 now?



Thank you for any response!

Shanghe Chen


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/jdk-dev/attachments/20240222/3cb06982/attachment-0001.htm>