Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Jiangli, Ok, I'll wait for the next webrev version. Thanks, Serguei On 7/8/16 18:04, Jiangli Zhou wrote: Hi Serguei, Thank you for looking into the changes. With Ioi and Karen’s suggestion for dynamically attached agent and -Xshare:on, I’m reworking the changes. I’m planning to remove t

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Ioi, > On Jul 8, 2016, at 5:47 PM, Ioi Lam wrote: > > Hi Jiangli, > > I agree what what Karen has said. > > If the user runs with -Xshare:on, but later attaches an agent, we just > pretend that the user had specified -Xshare:auto -- allow the agent to > receive all CFLH events, and stop l

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Serguei, Thank you for looking into the changes. With Ioi and Karen’s suggestion for dynamically attached agent and -Xshare:on, I’m reworking the changes. I’m planning to remove the check below and do that in a later place. I’ll post the new webrev next week. Thanks, Jiangli > On Jul 8, 20

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Jiangli, I agree what what Karen has said. If the user runs with -Xshare:on, but later attaches an agent, we just pretend that the user had specified -Xshare:auto -- allow the agent to receive all CFLH events, and stop loading from the shared archive. We should probably print out a warnin

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

On 7/8/16 12:59, serguei.spit...@oracle.com wrote: Hi Jiangli, src/share/vm/memory/metaspace.cpp 3177 if (UseSharedSpaces) { 3178 FileMapInfo* mapinfo = new FileMapInfo(); 3179 3180 if (JvmtiExport::should_post_class_file_load_hook()) { 3181 // Currently CDS does not support JVMTI CFHL when lo

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Jiangli, src/share/vm/memory/metaspace.cpp 3177 if (UseSharedSpaces) { 3178 FileMapInfo* mapinfo = new FileMapInfo(); 3179 3180 if (JvmtiExport::should_post_class_file_load_hook()) { 3181 // Currently CDS does not support JVMTI CFHL when loading shared class. 3182 // If JvmtiExport::should_p

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Karen, Thank you for the feedback. I think Ioi and your comments regarding the dynamically attached agent is very reasonable. I will rework the changes. Thanks, Jiangli > On Jul 8, 2016, at 7:55 AM, Karen Kinnear wrote: > > Jiangli, > > Thank you so much for picking up this bug fix, I ver

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Hi Ioi, Thanks for the comments. > On Jul 7, 2016, at 6:56 PM, Ioi Lam wrote: > > (Adding serviceability-dev@openjdk.java.net) > > Hi Jiangli, > > I think the two blocks of > >if (RequireSharedSpaces) { > tty->print_cr("An error has occurred while loading shared class."); > tty

RFR: 8160817:Add jsadebugd and classLoaderStat functionality to jhsdb

Hello, Please review the fix for: HYPERLINK "https://bugs.openjdk.java.net/browse/JDK-8160817"JDK-8160817 - Add jsadebugd and classLoaderStat functionality to jhsdb Webrev is at: http://cr.openjdk.java.net/~sballal/8160817/webrev.00/index.html This fix moves the jsadebugd and classLoader

Re: RFR: 8141341: CDS should be disabled if JvmtiExport::should_post_class_file_load_hook() is true

Jiangli, Thank you so much for picking up this bug fix, I very much appreciate that. I agree with not quitting the vm on agent attach. I actually think we also want the agent attach to succeed - see below. > On Jul 7, 2016, at 9:56 PM, Ioi Lam wrote: > > (Adding serviceability-dev@openjdk.java.

Re: CFV: New Serviceability Group Lead: Staffan Larsen

Vote: yes > On Jul 7, 2016, at 7:48 PM, Daniel D. Daugherty > wrote: > > I hereby nominate Staffan Larsen to Serviceability Group Lead [1]. > > Staffan is a Reviewer in the jdk9 and jdk8u Projects, a member of the > Serviceability team, and has so far committed > 145 changesets. He is the > Ser

Re: CFV: New Serviceability Group Lead: Staffan Larsen

Vote: yes