Re: [general] define pre-commit testing configs to gain the stability

Yes, It is reasonable and obvious. And I believe that committers are using it already. SY, Alexey 2006/10/10, Pavel Ozhdikhin <[EMAIL PROTECTED]>: On 10/9/06, Mikhail Loenko <[EMAIL PROTECTED]> wrote: > Hi Pavel, > > I'm sorry I did not catch how for example Nathan's commits will be checked > o

Re: [DRLVM][Threading/GC] thread info initialization problem

Before this bug is fixed in threading part, for any people who are interested in GCv5, please use patch below for workaround, just to add two lines. GCv5 only works in Windows at the moment. Please get it in latest SVN, and patch it with http://issues.apache.org/jira/browse/HARMONY-1785 for write

Re: [classlib][regex|luni] build break

Tim Ellison 写道: So I checked in a patch for HARMONY-688's regex fix, and it passed the regex unit tests, but causes the existing luni tests to fail in java.util.Scanner. I've not figured out the base cause of the failure so I've backed out the changes. Regards, Tim Hi regular expression gu

[DRLVM][Threading/GC] thread info initialization problem

Hi, when I tested with GCv5 in DRLVM, I found there is design issue in threading part. That is, after gc is initiated (gc_init) in runtime, the main thread can't allocate object based on its allocation context (thread local gc info), because the allocation context is not initialized (gc_thread_ini

Re: [classlib] [continuum] BUILD SUCCESSFUL: Classlib/linux.ia32 Build/Test

On 8 October 2006 at 5:38, Tim Ellison <[EMAIL PROTECTED]> wrote: > Mark Hindess wrote: > > FYI: I've changed the way our builds are reported to the -commits > > list. Now, the reports go to my apache.org email address where they are > > compressed and stored. The url in the message is then modi

[drlvm][jvmti] Last commit of Harmony-1582 brocks JVMTI support for DRLVM.

Hello, Fix for Harmony-1582 brocks initialization of JVMTI support. I'm investigating the problem and going to create fixing JIRA as soon as possible. P.S. May be include several JVMTI tests into "build test". -- Best regards, Pavel Rebriy

Re: [classlib][test] Failure in swing test

On 9 October 2006 at 16:02, Tim Ellison <[EMAIL PROTECTED]> wrote: > Works for me too now ... > > Unless anyone objects I suggest that we make the 'with.awt.swing' > default behavior (and we'll see the failure if it happens again). Excellent. I'll dump the property unless anyone else beats me t

Re: [DRLVM][JET] write barriers for C/C++-based GC

Good news is, I tried the write barrier implementation with today's SVN head, it runs with GCv5 perfectly. Well the running of GCv5 exposes some debug problem in DRLVM threading part. I need add two lines of code in GCv5 to work around the problem. I will start another thread to describe it. Tha

Re: [classlib][test] Failure in swing test

2006/10/10, Tim Ellison <[EMAIL PROTECTED]>: Works for me too now ... Unless anyone objects I suggest that we make the 'with.awt.swing' default behavior (and we'll see the failure if it happens again). Are you going to make it on Windows or both? I thought that linux dependencies are not curre

Re: [classlib][test] Failure in swing test

Nathan Beyer wrote: > +1 for defaulting with.awt.swing to true. > > Is there any reason to still have this property at all? Nope, I say we remove it Regards, Tim -- Tim Ellison ([EMAIL PROTECTED]) IBM Java technology centre, UK. ---

Re: [continuum] BUILD FAILURE: Classlib/linux.ia32 Build/Test

This failure is in RMI tests: The failing test is (my formatting) : Unexpected exception: java.rmi.ConnectIOException: Unable to acknowledge protocol with server; nested exception is: java.net.SocketTimeoutException: The operation timed out junit.framework.AssertionFailedError: Unexpected ex

Re: [drlvm] the big soup of VM properties (HARMONY-1626)

There are my thoughts about VM properties. I observe some problem with current VM properties implementation: 1) The current VM properties initialization is rather chaotic: * Some properties stored into two places: Properties class and array of VmStandardProperty structure. A certain algori

Re: [general] define pre-commit testing configs to gain the stability

On 10/9/06, Mikhail Loenko <[EMAIL PROTECTED]> wrote: Hi Pavel, I'm sorry I did not catch how for example Nathan's commits will be checked on the configurations he does not have? Since we have the problem with diversity of the hardware the committers own the following procedure might be reason

Re: [classlib] [testing] Coverage (was Re: 37% of total test execution time is spent in a single test)

On 10/6/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: > > - 'cc' - for cruise control script (and move current stuff to this dir); > > - 'coverage' - for coverage scripts (It will nice if somebody also placed > here data from issue 564); > > - 'japi' - for script to run 'japi'-tool. Agreed

Re: [classlib][pack200] Unpacked my first archive today!

Alex Blewitt wrote: I managed to unpack a Pack200 file completely today for the first time, so that by the end of it there were no bytes left in the queue. Unfortunately, it wasn't a spectacularly exciting archive -- just one interface with no methods (about as exciting as java.lang.Serializable,

Fwd: [icu-support] ICU4J & ICU4JNI 3.6 Release

Hello, If there is no objection, I will try to upgrade our nio charset provider to icu4jni 3.6. Best regards, Richard -- Forwarded message -- From: Ram Viswanadha <[EMAIL PROTECTED]> Date: Sep 30, 2006 8:22 AM Subject: [icu-support] ICU4J & ICU4JNI 3.6 Release To: [EMAIL PROTECT

Re: [classlib][regex|luni] build break

Spark Shen 写道: Tim Ellison 写道: So I checked in a patch for HARMONY-688's regex fix, and it passed the regex unit tests, but causes the existing luni tests to fail in java.util.Scanner. I've not figured out the base cause of the failure so I've backed out the changes. Regards, Tim Hi I wil

Re: [classlib][regex|luni] build break

Tim Ellison 写道: So I checked in a patch for HARMONY-688's regex fix, and it passed the regex unit tests, but causes the existing luni tests to fail in java.util.Scanner. I've not figured out the base cause of the failure so I've backed out the changes. Regards, Tim Hi I will look into the

Re: [classlib][test] Failure in swing test

+1 Nathan Beyer wrote: +1 for defaulting with.awt.swing to true. Is there any reason to still have this property at all? On 10/9/06, Tim Ellison <[EMAIL PROTECTED]> wrote: Works for me too now ... Unless anyone objects I suggest that we make the 'with.awt.swing' default behavior (and we'll s

Re: [classlib][regex|luni] build break

test package + 1 test package + 1 test package + 1 test package + 1 test package + 1 test package + 1 test package + 1 (no, I'm not volunteering) Tim Ellison wrote: So I checked in a patch for HARMONY-688's regex fix, and it passed the regex unit tests, but causes the existing luni tests to fai

Re: Can't get binary to work

Gregory Shimansky wrote: On Monday 09 October 2006 08:26 Armand Navabi wrote: Yes, I am still having problems. Like I said, I am just trying to run the executable currently. I see the same problem I was seeing when I built the DRLVM. I downloaded the Latest Linux JRE snapshot build, set the

Re: [drlvm][build]Stale object files are linked to DRLVM

Gregory Shimansky wrote: On Friday 06 October 2006 18:24 Salikh Zakirov wrote: I'm actually not. Were there an additional 24 hours in a day There is a whole list of reasons why I'm not a fan of the current system, including maintainability as well as performance. (Building classlib take

[classlib][regex|luni] build break

So I checked in a patch for HARMONY-688's regex fix, and it passed the regex unit tests, but causes the existing luni tests to fail in java.util.Scanner. I've not figured out the base cause of the failure so I've backed out the changes. Regards, Tim -- Tim Ellison ([EMAIL PROTECTED]) IBM Java

[DRLVM][JET] write barriers for C/C++-based GC

Thanks to Fursov and Alex for the write barrier submission. JIRA-1580 will be closed. Wish is committed as soon as possible that people can run GCv5 with it. http://issues.apache.org/jira/browse/HARMONY-1785 Thanks, xiaofeng On 10/9/06, Mikhail Fursov <[EMAIL PROTECTED]> wrote: To keep this t

Re: [continuum] BUILD FAILURE: Classlib/linux.ia32 Build/Test

I committed a regex fix, and it seems to have broken java.util.Scanner -- I'm looking... Tim [EMAIL PROTECTED] wrote: > Online report : > http://people.apache.org/~hindessm/continuum/classlib/linux.ia32/2006/10/09/20061009-234744.failure.log.gz > Build statistics: > State: Fa

Re: [classlib][test] Failure in swing test

+1 for defaulting with.awt.swing to true. Is there any reason to still have this property at all? On 10/9/06, Tim Ellison <[EMAIL PROTECTED]> wrote: Works for me too now ... Unless anyone objects I suggest that we make the 'with.awt.swing' default behavior (and we'll see the failure if it happ

Re: Can't get binary to work

On Monday 09 October 2006 08:26 Armand Navabi wrote: > Yes, I am still having problems. > > Like I said, I am just trying to run the executable currently. I see the > same problem I was seeing when I built the DRLVM. > > I downloaded the Latest Linux JRE snapshot build, set the JAVA_HOME, PATH > a

Re: [drlvm][build]Stale object files are linked to DRLVM

On Friday 06 October 2006 18:24 Salikh Zakirov wrote: > > I'm actually not. Were there an additional 24 hours in a day There > > is a whole list of reasons why I'm not a fan of the current system, > > including maintainability as well as performance. (Building classlib > > takes far less tim

Re: [classlib][test] Failure in swing test

Works for me too now ... Unless anyone objects I suggest that we make the 'with.awt.swing' default behavior (and we'll see the failure if it happens again). Regards, Tim Oleg Khaschansky wrote: > Tim, > > I can't see this failure at r454268. Anyone else see it? Maybe rebuild > will help? > > I

Re: [classlib][build] 'ant clean' failure

Vladimir Ivanov wrote: > Could somebody reproduce (and fix) the 'ant clean' failure for the > classlib? Done at r454492. Regards, Tim -- Tim Ellison ([EMAIL PROTECTED]) IBM Java technology centre, UK. - Terms of use : http://

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

I'm running debug build on WinXP, and I have following results so far: c_unit tests passed; smoke tests passed except gc.Los which hangs as usual; kernel tests works with the following success rates: Jet : 98.96% Opt: 99.15% Int : still running Thank you. Nik. On 10/9/06, Geir Magnusson

Re: [drlvm][gc/threading] need the two lsb's from object header for MMTk port

On 10/9/06, Salikh Zakirov <[EMAIL PROTECTED]> wrote: Weldon Washburn wrote: > Artem, > I'd like to confirm what you said. For GC_BIT_MASK, it looks like the > bottom > two bits of byte number one is available for exclusive GC use. If the GC > needs to alter these two bits while the mutator(s)

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

No it should not, but it is :). Look like there is no synchronization between setting interrupted flag and checking it, thus one may interrupt the thread between flag checking and actual wait, I will raise JIRA issue on this. Thank you. Nik. On 10/9/06, Evgueni Brevnov <[EMAIL PROTECTED]> wrot

Re: [drlvm] smoke test : gc PhantomReferenceQueueTest is really unstable

Geir, all, I did reserched this failure some time ago. The same failure was observed on gc.Finalizers. Here is what I've found. Now what is happening: 1) FinalizerThread is being run 2) java/lang/Object.notify()V is being compiled 3) java/lang/InternalError is being resolved (an

[drlvm] smoke test : gc PhantomReferenceQueueTest is really unstable

I get this more often than not. Can someone interested take a look? Happens w/ jitrino : java: /home/geir/dev/apache/harmony/enhanced/trunk/working_vm/vm/vmcore/src/class_support/C_Interface.cpp:524: const char* class_get_name(Class*): Assertion `cl' failed. SIGABRT in VM code. Stack trace:

[drlvm][jit][abcd] HARMONY-1788 Inequality graph generation from HIR

$subj Naveen, please, take a look! All, it is reasonable to commit the code since it becomes more and more difficult to support the patch queue. Now we have a mini-HOWTO for applying all the patches :) In default mode the code does not affect anyone. But still we can wait until the performance-

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

I will. On 10/9/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: Ok - I just committed 1582. Lets build on it... can someone test on winxp? I'll get to it in a bit, but would be happy to know sooner than later. Evgueni Brevnov wrote: > Ok, I see. Then it should be TM problem right? > > O

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

Ok - I just committed 1582. Lets build on it... can someone test on winxp? I'll get to it in a bit, but would be happy to know sooner than later. Evgueni Brevnov wrote: Ok, I see. Then it should be TM problem right? On 10/9/06, Nikolay Kuznetsov <[EMAIL PROTECTED]> wrote: Evgueni, acc

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

Ok, I see. Then it should be TM problem right? On 10/9/06, Nikolay Kuznetsov <[EMAIL PROTECTED]> wrote: Evgueni, according to the spec the thread wait method should throw interrupted exception even if it was interrupted before call. citing from the Object.wait method specification: "Throws

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

Evgueni, according to the spec the thread wait method should throw interrupted exception even if it was interrupted before call. citing from the Object.wait method specification: "Throws: InterruptedException - if another thread interrupted the current thread before or while the current thread

Re: [classlib][build]Lacks liblcms.a libpng.a and etc on Linux and recommend to add them to fetch-depends target

I see that you already said pretty much the same thing I did on another thread [1]. IMHO we should make sure that if the fetch-depends target finds it cannot sym link to preinstalled system libraries, then the user is clearly informed so they have the choice to install them without having the

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Evgueni Brevnov wrote: Geir, Salikh, cunit tests are very unstable.. that's why we have different results on different platforms. It all tests problems. I already fixed several places but it seems there is much more left. I believe we need to revise them all. The main problem is that sleep

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

at least cunit tests instabilities should not prevent the patch submission. It can take significant time to make them pass on different configurations. Evgueni On 10/9/06, Evgueni Brevnov <[EMAIL PROTECTED]> wrote: Geir, Salikh, cunit tests are very unstable.. that's why we have differ

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

that works. explain why Evgueni Brevnov wrote: Could you insert sleep_a_click(); just before line 59 right after hysem_create(&start, 0, 1): Does it help? On 10/9/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: [echo] ## Executing C unit test: test_ti_instrum [exec] Result: 1

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Geir, Salikh, cunit tests are very unstable.. that's why we have different results on different platforms. It all tests problems. I already fixed several places but it seems there is much more left. I believe we need to revise them all. The main problem is that sleep(somthing) is used to sync

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Could you insert sleep_a_click(); just before line 59 right after hysem_create(&start, 0, 1): Does it help? On 10/9/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: [echo] ## Executing C unit test: test_ti_instrum [exec] Result: 1 [echo] INFO: TEST test_jthread_get_all_threads st

Re: [drlvm][jitrino.JET] "-Xem jet:" and "-Xjit jet::wb4j" don't work anymore

To keep this thread up to date: the "WB for JET" code was commited to http://issues.apache.org/jira/browse/HARMONY-1785 by Alex. On 10/8/06, Mikhail Fursov <[EMAIL PROTECTED]> wrote: Ok I'll try to do it tomorrow. So you will have it on your monday's morning in JIRA On 10/8/06, Xiao-Feng Li <

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Geir Magnusson Jr. wrote: > > On Oct 9, 2006, at 9:58 AM, Salikh Zakirov wrote: > >> Geir Magnusson Jr. wrote: >>> I keep getting a failure when running the tests - >>> >>> test_jthread_get_all-threads failling the assertion at >>> test_ti_instrum.c:80 >>> >>> geir >> >> I have just tried to test

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

On Oct 9, 2006, at 10:38 AM, Evgueni Brevnov wrote: Geir, Did you pass tests on SUSE9? No, I don't have a SUSE9 machine. geir On 10/9/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: On Oct 9, 2006, at 9:58 AM, Salikh Zakirov wrote: > Geir Magnusson Jr. wrote: >> I keep getting a fai

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

[echo] ## Executing C unit test: test_ti_instrum [exec] Result: 1 [echo] INFO: TEST test_jthread_get_all_threads start [echo] ERROR: Assertion '(thread_count)==(i + initial_thread_count)' failed at /home/geir/dev/apache/harmony/ enhanced/trunk/working_vm/vm/tes ts/unit/threa

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Geir, Did you pass tests on SUSE9? On 10/9/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: On Oct 9, 2006, at 9:58 AM, Salikh Zakirov wrote: > Geir Magnusson Jr. wrote: >> I keep getting a failure when running the tests - >> >> test_jthread_get_all-threads failling the assertion at >> test_

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

On Oct 9, 2006, at 9:58 AM, Salikh Zakirov wrote: Geir Magnusson Jr. wrote: I keep getting a failure when running the tests - test_jthread_get_all-threads failling the assertion at test_ti_instrum.c:80 geir I have just tried to test HARMONY-1582 patch on Linux/ia32 SUSE 9 and Ubuntu 6

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Salikh Zakirov wrote: Geir Magnusson Jr. wrote: I keep getting a failure when running the tests - test_jthread_get_all-threads failling the assertion at test_ti_instrum.c:80 geir I have just tried to test HARMONY-1582 patch on Linux/ia32 SUSE 9 and Ubuntu 6. The smoke and unit tests both p

[drlvm] Extending ‘org.vmmagic.unboxed’ package with native calls and TLS access

All, We have 'org.vmmagic.unboxed' package (address arithmetic) support in both Jitrino.OPT and Jitrino.JET compilers today. So we can rewrite part of our VM helper's fast paths in Java and teach JIT compiler to inline it. The only thing is left to do is to get the answers to the following questio

Re: [classlib] build failure

Mikhail Loenko wrote: > Works fine on a fresh ws. > > could you please update up to 454342: > missed one file (was: moving HARMONY-1609 (Applet, ImageIO and Print > modules) files to correct locations) > > and try "ant rebuild" Verified. Classlib builds and works okay on revision r454349 cyrill

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

Geir Magnusson Jr. wrote: > I keep getting a failure when running the tests - > > test_jthread_get_all-threads failling the assertion at test_ti_instrum.c:80 > > geir I have just tried to test HARMONY-1582 patch on Linux/ia32 SUSE 9 and Ubuntu 6. The smoke and unit tests both pass okay with and

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

yes I see them too. I see them even w/o my patch :-( sure, it need to be fixed... On 10/9/06, Geir Magnusson Jr. <[EMAIL PROTECTED]> wrote: btw, why are there so many warnings? I'm not comfortable with having to suppress them - I'd rather fix them... On Oct 9, 2006, at 9:49 AM, Geir Magnusson

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

btw, why are there so many warnings? I'm not comfortable with having to suppress them - I'd rather fix them... On Oct 9, 2006, at 9:49 AM, Geir Magnusson Jr. wrote: sure - I'll redo the whole thing from clean... On Oct 9, 2006, at 9:45 AM, Evgueni Brevnov wrote: BTW, could you verify the

Re: [DRLVM][GC] first generational version of GCv5 is submitted

Cool! Thanks, -xiaofeng On 10/9/06, Weldon Washburn <[EMAIL PROTECTED]> wrote: My plan is to commit Xiao Feng's gcv5 and also build patches as soon as "build test" completes on windows and on Linux. This should happen in the next hour or so. We need to test the below command line invocation on

Re: [DRLVM][GC] first generational version of GCv5 is submitted

My plan is to commit Xiao Feng's gcv5 and also build patches as soon as "build test" completes on windows and on Linux. This should happen in the next hour or so. We need to test the below command line invocation once the above has happened. Its unlikely the above will disturb it. On 10/9/06,

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

sure - I'll redo the whole thing from clean... On Oct 9, 2006, at 9:45 AM, Evgueni Brevnov wrote: BTW, could you verify the right patch for tests was used? It should be tests.final.2.patch.. Sorry for the mess with patches Evgueni On 10/9/06, Evgueni Brevnov <[EMAIL PROTECTED]> wrote

Re: [drlvm] HARMONY-1582 - invocation API for DRLVM CHECKPOINT

BTW, could you verify the right patch for tests was used? It should be tests.final.2.patch.. Sorry for the mess with patches Evgueni On 10/9/06, Evgueni Brevnov <[EMAIL PROTECTED]> wrote: I put debug printing into test_ti_instrum.c and attached it to JIRA. Could you run it on your machi

Re: [drlvm] Jitrino.OPT performs incorrect GC enumeration in nested loop with array accesses

Agree with pluses and minuses :) 2 Egor, it is not so easy to find 'base' pointer having only interior pointer to an object. So, base pointer should be saved or it should be deducible somehow. Could someone review the changes in the patch to jitrino? I'm afraid I don't have sufficient knowledge

[classlib][build] 'ant clean' failure

Could somebody reproduce (and fix) the 'ant clean' failure for the classlib? Thanks, Vladimir <...> clean: [delete] Deleting: C:\harmony\classlib1.5\deploy\jdk\jre\lib\boot\x- net.jar [delete] Deleting 124 files from C:\harmony\classlib1.5\build\classes call-modules-one: -clean-kernel-patte

Re: [drlvm] Using latest linux snapshot[hdk]...

Yeah, I would have been surprised it that worked. Why not just install v 6? On Oct 9, 2006, at 9:18 AM, Dmitry Durnev wrote: Making symbolic link to libstdc++.so.5 not helps - I get just another error: Apache Harmony Launcher : (c) Copyright 1991, 2006 The Apache Software Foundation or its

Re: [classlib][build]Lacks liblcms.a libpng.a and etc on Linux and recommend to add them to fetch-depends target

I was thinking the same thing, but didn't think it all the way through... On Oct 9, 2006, at 9:11 AM, Oliver Deakin wrote: Can we provide archives of these dependencies in our svn (or wherever) as a fallback if fetch-depends cannot find them installed on the local system? Dont we do somet

Re: [general] define pre-commit testing configs to gain the stability

On Oct 9, 2006, at 8:52 AM, Mark Hindess wrote: On 9 October 2006 at 16:12, "Pavel Ozhdikhin" <[EMAIL PROTECTED]> wrote: On 10/9/06, Tim Ellison <[EMAIL PROTECTED]> wrote: Rana Dasgupta wrote: We need to check both release and debug builds...the binaries and timing characteristics are t

Re: [drlvm] the big soup of VM properties (HARMONY-1626)

On Oct 9, 2006, at 8:47 AM, Mikhail Fursov wrote: Here info about EM and JIT modules: 1) "org.apache.harmony.vm.vmdir" - EM uses this property to locate JIT libraries. 2) "vm.use_interpreter" - EM uses this property to detect if it's interpreter only mode. 3) "em.*" all EM properties. Actually

Re: [drlvm] Using latest linux snapshot[hdk]...

Making symbolic link to libstdc++.so.5 not helps - I get just another error: Apache Harmony Launcher : (c) Copyright 1991, 2006 The Apache Software Foundation or its licensors, as applicable. Failed to open JVM DLL: /export/workspace/snapshots/harmony-hdk-r450941/jdk/jre/bin/default/harmonyvm (/e

RE: [classlib][test] Site to collect classlib test statistics

Alexey, You've understood quiet right. The final goal is to point out bad patch, maybe I'll use bisect or other algorithm to find this commit. Anyway I need to collect statistics routinely each day to react fast! It's good idea to use results of cruise control not to do double work. -Origina

Re: [classlib][build]Lacks liblcms.a libpng.a and etc on Linux and recommend to add them to fetch-depends target

Can we provide archives of these dependencies in our svn (or wherever) as a fallback if fetch-depends cannot find them installed on the local system? Dont we do something like that on Windows at the moment? Regards, Oliver Leo Li wrote: Hi, Alexey: No I do not installed the developer versi

Re: [classlib/VMI] o.a.h.boot.class.path property

On Oct 9, 2006, at 6:32 AM, Alexey Varlamov wrote: I think we need to clarify VMI contract on "o.a.h.boot.class.path" property. Currently luni natives set this property to default bcp during JNI_OnLoad, accordingly to content of "%java.home%/lib/boot/bootclasspath.properties" file. Should the

Re: [classlib][test] Failure in swing test

Tim, I can't see this failure at r454268. Anyone else see it? Maybe rebuild will help? I also looked into SpinnerDateModelTest source and it doesn't seems to me that it could be related to the TransferHandler. -- Oleg On 10/9/06, Oleg Khaschansky <[EMAIL PROTECTED]> wrote: I will take a look

Re: [drlvm] Using latest linux snapshot[hdk]...

On the 0x1FD day of Apache Harmony Dmitry Durnev wrote: > When I try to run 'java -version' from the latest HDK snapshot, I get > the following error: > > Apache Harmony Launcher : (c) Copyright 1991, 2006 The Apache Software > Foundation or its licensors, as applicable. > Failed to open JVM DLL:

Re: [general] define pre-commit testing configs to gain the stability

On 9 October 2006 at 16:12, "Pavel Ozhdikhin" <[EMAIL PROTECTED]> wrote: > On 10/9/06, Tim Ellison <[EMAIL PROTECTED]> wrote: > > Rana Dasgupta wrote: > > > We need to check both release and debug builds...the binaries and timing > > > characteristics are too different. At this immediate stage of

Re: [drlvm] the big soup of VM properties (HARMONY-1626)

Here info about EM and JIT modules: 1) "org.apache.harmony.vm.vmdir" - EM uses this property to locate JIT libraries. 2) "vm.use_interpreter" - EM uses this property to detect if it's interpreter only mode. 3) "em.*" all EM properties. Actually there are only two today: a) " em.properties" - the

Re: [classlib] build failure

yup, Ive just updated and rebuilt, and it works fine for me too. Mikhail Loenko wrote: Works fine on a fresh ws. could you please update up to 454342: missed one file (was: moving HARMONY-1609 (Applet, ImageIO and Print modules) files to correct locations) and try "ant rebuild" Thanks, Mikha

Re: [testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

BTW, it seems Thread.interrupt() doesn't clear thread's interrupt status as spec states. Could we add such test cases as well Evgueni On 10/9/06, Evgueni Brevnov <[EMAIL PROTECTED]> wrote: Hi, While testing HARMONY-1582 (invocation api) it turned out that one kernel test starts to fail wit

[testing][kernel.tests] j.l.Thread test fails on HARMONY-1582

Hi, While testing HARMONY-1582 (invocation api) it turned out that one kernel test starts to fail with the following message: There were 3 failures: 1) testInterrupt_Joining(java.lang.ThreadTest)junit.framework.AssertionFailed Error: joining thread has not received the InterruptedException

Re: [classlib] build failure

Works fine on a fresh ws. could you please update up to 454342: missed one file (was: moving HARMONY-1609 (Applet, ImageIO and Print modules) files to correct locations) and try "ant rebuild" Thanks, Mikhail 2006/10/9, Mikhail Loenko <[EMAIL PROTECTED]>: I couldn't reproduce that - worked fin

[drlvm] Using latest linux snapshot[hdk]...

When I try to run 'java -version' from the latest HDK snapshot, I get the following error: Apache Harmony Launcher : (c) Copyright 1991, 2006 The Apache Software Foundation or its licensors, as applicable. Failed to open JVM DLL: /export/workspace/snapshots/harmony-hdk-r450941/jdk/jre/bin/default

Re: [classlib] build failure

I couldn't reproduce that - worked fine before I committed Now trying on a fresh ws 2006/10/9, Alexey Petrenko <[EMAIL PROTECTED]>: Yes, this happens because HARMONY-1729 was not integrated yet. HARMONY-1729 consists of build files diff and connetced to HARMONY-1609. SY, Alexey 2006/10/9, Sa

Re: [classlib] build failure

Yes, this happens because HARMONY-1729 was not integrated yet. HARMONY-1729 consists of build files diff and connetced to HARMONY-1609. SY, Alexey 2006/10/9, Salikh Zakirov <[EMAIL PROTECTED]>: Hi, I cannot build classlib on the latest revision ([r454318] moving HARMONY-1609 (Applet, ImageIO

Re: [drlvm] Non-bug difference HARMONY-1688?

2006/10/9, Alexey Varlamov <[EMAIL PROTECTED]>: 2006/10/9, Alexey Varlamov <[EMAIL PROTECTED]>: > 2006/10/9, Pavel Pervov <[EMAIL PROTECTED]>: > > Commenting on (1) I should note, that VM generally can't determine the fact > > of "initiating" class loading; it can only be done in Java (except boo

Re: [drlvm] Non-bug difference HARMONY-1688?

2006/10/9, Alexey Varlamov <[EMAIL PROTECTED]>: 2006/10/9, Pavel Pervov <[EMAIL PROTECTED]>: > Commenting on (1) I should note, that VM generally can't determine the fact > of "initiating" class loading; it can only be done in Java (except bootstrap > class loader, of course). I presume you mean

Re: [drlvm] Non-bug difference HARMONY-1688?

2006/10/9, Pavel Pervov <[EMAIL PROTECTED]>: Commenting on (1) I should note, that VM generally can't determine the fact of "initiating" class loading; it can only be done in Java (except bootstrap class loader, of course). I presume you mean cases of direct calls to loadClass() from executed Ja

[classlib] build failure

Hi, I cannot build classlib on the latest revision ([r454318] moving HARMONY-1609 (Applet, ImageIO and Print modules) fil es to) with the following errors: [javac] /export/users/salikh/classlib/trunk/modules/accessibility/src/main/java/javax/accessibility/AccessibleRelationSet.java:31: err

Re: [drlvm] Jitrino.OPT performs incorrect GC enumeration in nested loop with array accesses

On 10/9/06, Salikh Zakirov <[EMAIL PROTECTED]> wrote: IMHO, the best solution would be to use (slot,offset) everywhere, and pre-compute and cache offsets before enumerating the base pointer. +1. Already did it in the patch. Or even enumerate base pointer *after* all interior pointers that de

Re: [drlvm] Jitrino.OPT performs incorrect GC enumeration in nested loop with array accesses

On 09 Oct 2006 14:25:58 +0700, Egor Pasko <[EMAIL PROTECTED]> wrote: BTW, what is mptr? is it 'base + offset' by definition? today I am more about dummy questions :) 'offset' is sometimes an operand (a mem location or a register) and not a mere int. As in our example, when arrays are accessed.

Re: [drlvm] Non-bug difference HARMONY-1688?

Commenting on (1) I should note, that VM generally can't determine the fact of "initiating" class loading; it can only be done in Java (except bootstrap class loader, of course). AFAIK, there is the bug: DRLVM does not record initiating class loader of a class. (1) is just the consequence. On 10

Re: [drlvm] Jitrino.OPT performs incorrect GC enumeration in nested loop with array accesses

Mikhail Fursov wrote: > On 10/7/06, Weldon Washburn <[EMAIL PROTECTED]> wrote: >> >> #3 might have originally been put in the JIT/GC interface because the JIT >> developers liked it. I can't think of a reason why a GC would prefer >> this >> interface over #2 above. > > If it could be JIT develo

Re: [drlvm] Jitrino.OPT performs incorrect GC enumeration in nested loop with array accesses

On 10/6/06, Ivan Volosyuk <[EMAIL PROTECTED]> wrote: If it is possible to get somehow valid offset within object then it is great! I would be happy to see this solved this way. Waiting for your patch. Folks, I put the patch "jit_gc.diff

Re: [classlib][test] Failure in swing test

I will take a look at this. On 10/8/06, Tim Ellison <[EMAIL PROTECTED]> wrote: I see a failure on IA32 Win XP tests at r454168 (after applying the TransferHandler patch). The walkback is: junit.framework.AssertionFailedError: expected:<0> but was:<7> at javax.swing.SpinnerDateModelTest.test

[classlib/VMI] o.a.h.boot.class.path property

I think we need to clarify VMI contract on "o.a.h.boot.class.path" property. Currently luni natives set this property to default bcp during JNI_OnLoad, accordingly to content of "%java.home%/lib/boot/bootclasspath.properties" file. Should the VM sync value of this property back with added path to

[drlvm] Non-bug difference HARMONY-1688?

Folks, Would you mind to spend some time looking at HARMONY-1688? Good knowledge of classloading and security machinery is an advantage :) In short, we face 2 issues in DRLVM: 1) DRLVM itself is not aware of classes initiated (in contrast to defined ones) by a loader, and always delegates to Jav

Re: [classlib][pack200] Unpacked my first archive today!

Wow wow! My congratulations. Don't stop on this. :) 2006/10/6, Alex Blewitt <[EMAIL PROTECTED]>: I managed to unpack a Pack200 file completely today for the first time, so that by the end of it there were no bytes left in the queue. Unfortunately, it wasn't a spectacularly exciting archive -- ju

Re: [DRLVM][GC] first generational version of GCv5 is submitted

Salikh, right. I've submitted a patch to support both GC build. Hopefully Weldon will commit it today. Thanks, -xiaofeng On 10/9/06, Salikh Zakirov <[EMAIL PROTECTED]> wrote: Weldon Washburn wrote: > On second thought, you are right. I don't really should not be hacking on > the framework rig

Re: [drlvm][gc/threading] need the two lsb's from object header for MMTk port

Weldon Washburn wrote: > Artem, > I'd like to confirm what you said. For GC_BIT_MASK, it looks like the > bottom > two bits of byte number one is available for exclusive GC use. If the GC > needs to alter these two bits while the mutator(s) are running, the GC > needs > to use CAS on the entire 32

Re: [classlib] enabling AWT/Swing by default

No objections to this "swing-only" solution. Thanks, 2006/10/6, Oleg Khaschansky <[EMAIL PROTECTED]>: Tim, I attached a patch which doesn't have side effects to HARMONY-1723 :) -- Oleg On 10/6/06, Tim Ellison <[EMAIL PROTECTED]> wrote: > Oleg Khaschansky wrote: > >> So what happens to the pa

Re: [jira] Updated: (HARMONY-1763) [classlib][beans] method java.beans.PropertyDescriptor.equals(Object) throws ClassCastException

Alexei, Thank you for paying your attention to this. The patch is already applied, sorry. Anyway, the suggested code is not significantly different, IMO. Actually, I tend to make minimal modifications to the original code when possible. On 10/6/06, Alexei Zakharov <[EMAIL PROTECTED]> wrote: Ol

  1   2   >