Re: RFR(L) 8237354: Add option to jcmd to write a gzipped heap dump

Hi Ralf, On 12/02/2020 6:41 pm, Schmelter, Ralf wrote: Hi David, I see little point in subclassing NonJavaThread (via NamedThread) but then overriding pre_run() and post_run() so that you don't do anything that NonJavaThread is supposed to do regarding the NJT iterator capabilities. The prob

Re: RFR: JDK-8238710: LingeredApp doesn't log stdout/stderr if exits with non-zero code

Hi Alex, LGTM++ Thanks, Serguei On 2/12/20 3:45 PM, Chris Plummer wrote: Ok. LGTM. Chris On 2/12/20 1:58 PM, Alex Menkov wrote: Hi Chris, thanks for the review. finishApp is also called from startApp(String... cmd) method and appProcess can be not initialized there. In the case finishApp w

Re: RFR: 8238196: tests that use SA Attach should not be allowed to run against signed binaries on Mac OS X 10.14.5 and later

yes, I'm fine w/ that. -- Igor > On Feb 12, 2020, at 3:48 PM, Chris Plummer wrote: > > On 2/12/20 2:15 PM, David Holmes wrote: >> On 13/02/2020 5:02 am, Chris Plummer wrote: >>> Hi David, >>> What you have below is a mix of #2 and #3 - you convert to a generic exception but also re-

Re: RFR: 8238196: tests that use SA Attach should not be allowed to run against signed binaries on Mac OS X 10.14.5 and later

On 2/12/20 2:15 PM, David Holmes wrote: On 13/02/2020 5:02 am, Chris Plummer wrote: Hi David, What you have below is a mix of #2 and #3 - you convert to a generic exception but also re-assert the interrupt state. That's a little unusual. That what I also thought which is why I was suggesting

Re: RFR: JDK-8238710: LingeredApp doesn't log stdout/stderr if exits with non-zero code

Ok. LGTM. Chris On 2/12/20 1:58 PM, Alex Menkov wrote: Hi Chris, thanks for the review. finishApp is also called from startApp(String... cmd) method and appProcess can be not initialized there. In the case finishApp will throw NPE (calling appProcess.exitValue()) --alex On 02/12/2020 13:53,

Re: RFR: 8238196: tests that use SA Attach should not be allowed to run against signed binaries on Mac OS X 10.14.5 and later

On 13/02/2020 5:02 am, Chris Plummer wrote: Hi David, What you have below is a mix of #2 and #3 - you convert to a generic exception but also re-assert the interrupt state. That's a little unusual. That what I also thought which is why I was suggesting not doing the interrupt() call and only

Re: RFR: JDK-8238710: LingeredApp doesn't log stdout/stderr if exits with non-zero code

Hi Chris, thanks for the review. finishApp is also called from startApp(String... cmd) method and appProcess can be not initialized there. In the case finishApp will throw NPE (calling appProcess.exitValue()) --alex On 02/12/2020 13:53, Chris Plummer wrote: Hi Alex, Thanks for doing this. Not

Re: RFR: JDK-8238710: LingeredApp doesn't log stdout/stderr if exits with non-zero code

Hi Alex, Thanks for doing this. Not having output from a spawned process that failed is an issue with more than just LingeredApp tests. This is a good start in getting those fixed. I'm a little unclear on one part of your fix. Why did you move the "appProcess != null" into finishApp(). You a

RFR: JDK-8238710: LingeredApp doesn't log stdout/stderr if exits with non-zero code

Hi all, Please review small fix for https://bugs.openjdk.java.net/browse/JDK-8238710 webrev: http://cr.openjdk.java.net/~amenkov/jdk15/LingeredApp_log_error/webrev/ --alex

Re: [PING] RFR: 8231111: Cgroups v2: Rework Metrics in java.base so as to recognize unified hierarchy

I'll take a look next couple days.  I was out last few days and am catching up on other things. Mandy On 2/11/20 10:04 AM, Severin Gehwolf wrote: Hi Mandy, Bob, Thanks again for the reviews and patience on this. Sorry it took me so long to get back to this :-/ Updated webrev: Full: http://cr

Re: RFR: 8238196: tests that use SA Attach should not be allowed to run against signed binaries on Mac OS X 10.14.5 and later

Hi David, What you have below is a mix of #2 and #3 - you convert to a generic exception but also re-assert the interrupt state. That's a little unusual. That what I also thought which is why I was suggesting not doing the interrupt() call and only throwing the RuntimeException. I agree that

Re: [PING] RFR: 8231111: Cgroups v2: Rework Metrics in java.base so as to recognize unified hierarchy

Hi, On Wed, 2020-02-12 at 12:29 -0500, Bob Vandette wrote: > I applied the delegation change that you recommended and now all container > tests pass. > > Here’s the change that I applied. > > echo +cpu +cpuset +io +memory +pids > /sys/fs/cgroup/cgroup.subtree_control > echo +cpu +cpuset +io +me

Re: [PING] RFR: 8231111: Cgroups v2: Rework Metrics in java.base so as to recognize unified hierarchy

I applied the delegation change that you recommended and now all container tests pass. Here’s the change that I applied. echo +cpu +cpuset +io +memory +pids > /sys/fs/cgroup/cgroup.subtree_control echo +cpu +cpuset +io +memory +pids > /sys/fs/cgroup/user.slice/cgroup.subtree_control echo +cpu +

Re: RFR(L) 8237354: Add option to jcmd to write a gzipped heap dump

On 2/12/20 4:17 AM, Schmelter, Ralf wrote: Hi Serguei, the use case is being able to get a heap dump from big Java servers. These usually run on machines with a lot of memory and CPUs, but not much disk space (which they don't need apart from some trace files and the server code itself). A

Re: [PING] RFR: 8231111: Cgroups v2: Rework Metrics in java.base so as to recognize unified hierarchy

> On Feb 12, 2020, at 6:22 AM, Severin Gehwolf wrote: > > Hi Bob, > > On Tue, 2020-02-11 at 16:59 -0500, Bob Vandette wrote: >> I applied your patch to the latest JDK 15 sources and ran the container >> tests on Oracle Linux 8.1 with podman/cgroupv2 enabled. There were some >> issues. >> I’m

Re: RFR: 8238196: tests that use SA Attach should not be allowed to run against signed binaries on Mac OS X 10.14.5 and later

Hi Chris, I think you are overthinking this. :) What you have observed is that the code that actually uses this method does not utilise interrupts, or expect them, so if you artifically inject one in this library method then you see things failing in unexpected ways. That also means that if t

RE: RFR(L) 8237354: Add option to jcmd to write a gzipped heap dump

Hi Serguei, the use case is being able to get a heap dump from big Java servers. These usually run on machines with a lot of memory and CPUs, but not much disk space (which they don't need apart from some trace files and the server code itself). And if we can get the customer to mount some NFS

Re: [PING] RFR: 8231111: Cgroups v2: Rework Metrics in java.base so as to recognize unified hierarchy

Hi Bob, On Tue, 2020-02-11 at 16:59 -0500, Bob Vandette wrote: > I applied your patch to the latest JDK 15 sources and ran the container > tests on Oracle Linux 8.1 with podman/cgroupv2 enabled. There were some > issues. > I’m not sure if its my setup or not. Looking over them, they appear to

RFR(S) 8238585: Use handshake for JvmtiEventControllerPrivate::enter_interp_only_mode() and don't make compiled methods on stack not_entrant

// Repost including hotspot runtime and gc lists. // Dean Long suggested to do so, because the enhancement replaces a vm operation // with a handshake. // Original thread: http://mail.openjdk.java.net/pipermail/serviceability-dev/2020-February/030359.html Hi, could I please get reviews for this

RE: RFR(S) 8238585: Use handshake for JvmtiEventControllerPrivate::enter_interp_only_mode() and don't make compiled methods on stack not_entrant

Ok. I will repost and include hotspot runtime and gc lists. Thanks, Richard. -Original Message- From: Dean Long Sent: Dienstag, 11. Februar 2020 18:28 To: Reingruber, Richard ; serviceability-dev@openjdk.java.net; hotspot-compiler-...@openjdk.java.net Subject: Re: RFR(S) 8238585: Use h

RE: RFR(L) 8237354: Add option to jcmd to write a gzipped heap dump

Hi Serguei, the problem is, that if I make the -gz flag a jlong argument, I cannot just use '-gz'. This causes DCmdArgument::parse_value() to be called with a NULL string, which leads to an error. That is why a used a string argument in my code. But using a string when I really mean an integer

RE: RFR(L) 8237354: Add option to jcmd to write a gzipped heap dump

Hi David, > I see little point in subclassing NonJavaThread (via NamedThread) but > then overriding pre_run() and post_run() so that you don't do anything > that NonJavaThread is supposed to do regarding the NJT iterator > capabilities. The problem is the post_run() method of NamedThread call

Re: [15] RFR 8238633: JVMTI heap walk should consult GC for marking oops

Hi Zhengyu, On 2020-02-07 16:53, Zhengyu Gu wrote: Hi, I would like purpose this change that allows GC to provide ObjectMarker during JVMTI heap walk. Currently, JVMTI heap walk uses oop markword's 'marked' pattern to indicate 'visited' oop. Unfortunately, it conflicts with Shenandoah, wh

Re: RFR: 8238196: tests that use SA Attach should not be allowed to run against signed binaries on Mac OS X 10.14.5 and later

Hi Igor, I think it might be best to the interrupt() call out. I wanted to see what would happen if we ever got an InterruptedException, so I added the following to the start of Platform.shouldSAAttach():     try {     throw new Int