Mailing-List: contact commits-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Message-ID: <18930332.1173769232721.JavaMail.jira@brutus>
Date: Tue, 13 Mar 2007 00:00:32 -0700 (PDT)
From: "Nina Rinskaya (JIRA)" <jira@apache.org>
To: commits@harmony.apache.org
Subject: [jira] Commented: (HARMONY-3219) [drlvm][jit] Eclipse Unit Tests'
 suite crashes on Linux since r502549
In-Reply-To: <30001519.1172070485785.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HARMONY-3219?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12480291 ] 

Nina Rinskaya commented on HARMONY-3219:
----------------------------------------

Yes, I can't reproduce the issue either (r517220).

> [drlvm][jit] Eclipse Unit Tests' suite crashes on Linux since r502549
> ---------------------------------------------------------------------
>
>                 Key: HARMONY-3219
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3219
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: Linux x86 (SLES9)
>            Reporter: Nina Rinskaya
>         Attachments: linux_x86_default_options.txt, linux_x86_HARMONY-3041_reverted_default_options.txt, linux_x86_opt.txt
>
>
> Looks that r502549 caused EUT regression: "ui" testsuite crashes on Linux. The issue is not reproducible with -Xem:jet, -Xint. Another crash on the same test occurs when running with -Xem:opt (see stacktrace attached).
> Several days ago I was able to revert to r502548/r502549 and check that the issue is not reproducible with r502548, but now I can't build classlib&vm when reverted to these revisions.
> Moreover, if I revert HARMONY-3041 (which is r502549 commit), it now fails on a different way (see atachment).
> To reproduce:
> Download scripts from http://issues.apache.org/jira/browse/HARMONY-2038, unpack archive,
> specify valid settings in eut.properties, run 
> >ant setup
> and then
> >ant -Dtests=ui >log 2>&1
> (see readme.txt for details)
> Then see log file for org.eclipse.ui.tests.UiTestSuite sub-suite.
> Stacktraces summary (complete stacktraces are attached):
> Linux x86 (SLES9):
> Default options (none):
> SIGSEGV in VM code
>        1: array_copy(ManagedObject*, int, ManagedObject*, int, int)
> (/export/users/nrinskay/EUT_CC/trunk/cc/p
> rojects/drlvm/trunk/vm/vmcore/src/object/vm_arrays.cpp:433)
>        2: array_copy_jni
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/vmcore/src/jni/jni_u
> tils.cpp:637)
>        3: Java_java_lang_VMMemoryManager_arrayCopy
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/tru
> nk/vm/vmcore/src/kernel_classes/native/java_lang_VMMemoryManager.cpp:48)
>        4: IP is 0x83F2E9F1 <native code>
>        5: java/lang/System.arraycopy(Ljava/lang/Object;ILjava/lang/Object;II)V
> (System.java:86)
>        6: org/eclipse/swt/custom/CTabFolder.getItems()[Lorg/eclipse/swt/custom/CTabItem;
> (CTabFolder.java:118
> 9)
> -------------------------------
> Linux x86, -Xem:opt:
>    [exec]      [java] java:
> /export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/gc_types.h:197:
> void Partial_Reveal_Object::valid(): As
> sertion `class_get_allocation_handle(c) == vt()' failed.
>     [exec]      [java] SIGABRT in VM code.
>     [exec]      [java] Stack trace:
>     [exec]      [java]         1: ?? (??:-1)
>     [exec]      [java]         2: abort (??:-1)
>     [exec]      [java]         3: __assert_fail (??:-1)
>     [exec]      [java]         4: Partial_Reveal_Object::valid()
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/gc_types.h:198)
>     [exec]      [java]         5: mark_object(Partial_Reveal_Object*)
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/collect_slide
> _compact.cpp:106)
>     [exec]      [java]         6: slide_process_object
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/collect_slide_compact.cpp:50
> 3)
>     [exec]      [java]         7: gc_slide_add_root_set_entry(Slot)
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/collect_slide_c
> ompact.cpp:571)
>     [exec]      [java]         8: gc_add_root_set_entry_slot(Slot)
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/slot.cpp:252)
>     [exec]      [java]         9: gc_add_root_set_entry
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/gc_cc/src/slot.cpp:263)
>     [exec]      [java]         10: vm_enumerate_root_reference
> (/export/users/nrinskay/EUT_CC/trunk/cc/projects/drlvm/trunk/vm/vmcore/src/gc/root_set_enum_co
> mmon.cpp:139)
>  .............
> ---------------------------------
> With HARMONY-3041 reverted:
>   [exec]      [java] SIGSEGV in VM code.
>     [exec]      [java] Stack trace:
>     [exec]      [java]         1:
> Jitrino::MemoryManager::alloc(unsigned int) (??:-1)
>     [exec]      [java]         2:
> Jitrino::StlMMAllocator<Jitrino::Ia32::SpillGen::Op>::allocate(unsigned
> int, void const*) (??:-1)
>     [exec]      [java]         3:
> std::_Vector_alloc_base<Jitrino::Ia32::SpillGen::Op,
> Jitrino::StlMMAllocator<Jitrino::Ia32::SpillGen::Op>, false>::_M_alloc
> ate(unsigned int) (??:-1)
>     [exec]      [java]         4: ?? (002d7911
>     [exec]      [java] :2)
>     [exec]      [java]         5:
> std::vector<Jitrino::Ia32::SpillGen::Op,
> Jitrino::StlMMAllocator<Jitrino::Ia32::SpillGen::Op>
> >::push_back(Jitrino::Ia32::S
> pillGen::Op const&) (??:-1)
>     [exec]      [java]         6:
> Jitrino::Ia32::SpillGen::Opline::addOp(Jitrino::Ia32::SpillGen::Instx*,
> char) (??:-1)
>     [exec]      [java]         7: Jitrino::Ia32::SpillGen::pass0() (??:-1)
>     [exec]      [java]         8: Jitrino::Ia32::SpillGen::runImpl() (??:-1)
>     [exec]      [java]         9: Jitrino::Ia32::SessionAction::run() (??:-1)
>     [exec]      [java]         10:
> Jitrino::runPipeline(Jitrino::CompilationContext*) (??:-1)
> ...............
> -------------------------------------
> Please find detailed stacktaces attached.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.