[rvm-research] DaCapo:chart is broken?
|
View:
New views
11 Messages
—
Rating Filter:
Alert me
|
 |
[rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
Hi,
I had checked out JikesRVM r15150 (HEAD today) and built it with: $ ant very-clean && bin/buildit --nuke localhost prototype after that, I'm trying to run DaCapo:chart on Gentoo x86, and getting the failure: $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10-MR2.jar chart ===== DaCapo chart starting ===== Fatal error: NullPointerException within uninterruptible region. Exiting virtual machine due to uninterruptibility violation. -- Stack -- at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 at Lorg/jikesrvm/runtime/RuntimeEntrypoints; deliverHardwareException(II)V at line 685 at <hardware trap> at Lorg/jikesrvm/runtime/Magic; getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; at line 166 at Lorg/jikesrvm/runtime/RuntimeEntrypoints; unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; at line 1003 at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at line 61 at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 at Lorg/jikesrvm/classloader/RVMClass; getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 at Lorg/jikesrvm/jni/JNIFunctions; FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/Address;)I at line 185 at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> JikesRVM: internal error: recursive use of hardware exception registers (exiting) Scheduler.dumpStack(): in a recursive call, 2 deep. -- Stack -- at Lorg/jikesrvm/scheduler/Scheduler; dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/Address;)V at line 732 at Lorg/jikesrvm/scheduler/Scheduler; dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at line 650 at Lorg/jikesrvm/scheduler/Scheduler; traceback(Ljava/lang/String;)V at line 628 at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 at Lorg/jikesrvm/runtime/RuntimeEntrypoints; deliverHardwareException(II)V at line 685 at <hardware trap> at Lorg/jikesrvm/runtime/Magic; getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; at line 166 at Lorg/jikesrvm/runtime/RuntimeEntrypoints; unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; at line 1003 at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at line 61 at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 at Lorg/jikesrvm/classloader/RVMClass; getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 at Lorg/jikesrvm/jni/JNIFunctions; FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/Address;)I at line 185 at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> at <native frame> JikesRVM: internal error: recursive use of hardware exception registers (exiting) I also checked other Dacapo benches and they all were finished correctly on this prototype configuration. This issue also reproduces for me at r15100, r15000, r14950. The cattrack is down, so I don't know whether this issue is reproduced on test machines. JIRA search for "Dacapo chart" didn't bring me to any of relevant issues. Is this new or some known issue? Thanks, Aleksey. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Ian Rogers
::
Rate this Message:
Hi Aleksey,
cattrack is back up again. I wasn't aware of any chart regressions, but it isn't executed as part of pre-commit due to its dependence on gtk-peers in Classpath (we don't build any peer sets on x86 64, on IA32 we do build a peer set iirc but it will only work if it sees a valid X server). Regards, Ian 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> Hi, ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
Thanks, Ian.
Cattrack didn't catch any regressions on latest sanity check [1]. It looks like the problem is machine-dependent. Where can I get the info on exact HW/SW used on exact server (I'm interested in jikesrvm.anu.edu.au)? P.S. Cattrack seems to die at my hands... again. Thanks, Aleksey. [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers <rogers.email@...> wrote: > Hi Aleksey, > > cattrack is back up again. I wasn't aware of any chart regressions, but it > isn't executed as part of pre-commit due to its dependence on gtk-peers in > Classpath (we don't build any peer sets on x86 64, on IA32 we do build a > peer set iirc but it will only work if it sees a valid X server). > > Regards, > Ian > > 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >> >> Hi, >> >> I had checked out JikesRVM r15150 (HEAD today) and built it with: >> >> $ ant very-clean && bin/buildit --nuke localhost prototype >> >> after that, I'm trying to run DaCapo:chart on Gentoo x86, and getting >> the failure: >> >> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10-MR2.jar chart >> >> ===== DaCapo chart starting ===== >> >> Fatal error: NullPointerException within uninterruptible region. >> Exiting virtual machine due to uninterruptibility violation. >> >> -- Stack -- >> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> deliverHardwareException(II)V at line 685 >> at <hardware trap> >> at Lorg/jikesrvm/runtime/Magic; >> >> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >> at line 166 >> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> >> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >> at line 1003 >> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at line 61 >> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >> at Lorg/jikesrvm/classloader/RVMClass; >> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >> at Lorg/jikesrvm/jni/JNIFunctions; >> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/Address;)I >> at line 185 >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> JikesRVM: internal error: recursive use of hardware exception >> registers (exiting) >> Scheduler.dumpStack(): in a recursive call, 2 deep. >> >> -- Stack -- >> at Lorg/jikesrvm/scheduler/Scheduler; >> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/Address;)V >> at line 732 >> at Lorg/jikesrvm/scheduler/Scheduler; >> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at line >> 650 >> at Lorg/jikesrvm/scheduler/Scheduler; >> traceback(Ljava/lang/String;)V at line 628 >> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> deliverHardwareException(II)V at line 685 >> at <hardware trap> >> at Lorg/jikesrvm/runtime/Magic; >> >> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >> at line 166 >> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> >> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >> at line 1003 >> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at line 61 >> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >> at Lorg/jikesrvm/classloader/RVMClass; >> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >> at Lorg/jikesrvm/jni/JNIFunctions; >> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/Address;)I >> at line 185 >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> at <native frame> >> JikesRVM: internal error: recursive use of hardware exception >> registers (exiting) >> >> I also checked other Dacapo benches and they all were finished >> correctly on this prototype configuration. This issue also reproduces >> for me at r15100, r15000, r14950. The cattrack is down, so I don't >> know whether this issue is reproduced on test machines. JIRA search >> for "Dacapo chart" didn't bring me to any of relevant issues. >> >> Is this new or some known issue? >> >> Thanks, >> Aleksey. >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the >> world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Jikesrvm-researchers mailing list >> Jikesrvm-researchers@... >> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Jikesrvm-researchers mailing list > Jikesrvm-researchers@... > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
> (I'm interested in jikesrvm.anu.edu.au)?
Argh, sorry, I meant jalapeno.anu.edu.au. On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev <aleksey.shipilev@...> wrote: > Thanks, Ian. > > Cattrack didn't catch any regressions on latest sanity check [1]. It > looks like the problem is machine-dependent. Where can I get the info > on exact HW/SW used on exact server (I'm interested in > jikesrvm.anu.edu.au)? > > P.S. Cattrack seems to die at my hands... again. > > Thanks, > Aleksey. > > [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart > > On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers <rogers.email@...> wrote: >> Hi Aleksey, >> >> cattrack is back up again. I wasn't aware of any chart regressions, but it >> isn't executed as part of pre-commit due to its dependence on gtk-peers in >> Classpath (we don't build any peer sets on x86 64, on IA32 we do build a >> peer set iirc but it will only work if it sees a valid X server). >> >> Regards, >> Ian >> >> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >>> >>> Hi, >>> >>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >>> >>> $ ant very-clean && bin/buildit --nuke localhost prototype >>> >>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and getting >>> the failure: >>> >>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10-MR2.jar chart >>> >>> ===== DaCapo chart starting ===== >>> >>> Fatal error: NullPointerException within uninterruptible region. >>> Exiting virtual machine due to uninterruptibility violation. >>> >>> -- Stack -- >>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>> deliverHardwareException(II)V at line 685 >>> at <hardware trap> >>> at Lorg/jikesrvm/runtime/Magic; >>> >>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >>> at line 166 >>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>> >>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >>> at line 1003 >>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at line 61 >>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>> at Lorg/jikesrvm/classloader/RVMClass; >>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>> at Lorg/jikesrvm/jni/JNIFunctions; >>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/Address;)I >>> at line 185 >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> JikesRVM: internal error: recursive use of hardware exception >>> registers (exiting) >>> Scheduler.dumpStack(): in a recursive call, 2 deep. >>> >>> -- Stack -- >>> at Lorg/jikesrvm/scheduler/Scheduler; >>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/Address;)V >>> at line 732 >>> at Lorg/jikesrvm/scheduler/Scheduler; >>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at line >>> 650 >>> at Lorg/jikesrvm/scheduler/Scheduler; >>> traceback(Ljava/lang/String;)V at line 628 >>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>> deliverHardwareException(II)V at line 685 >>> at <hardware trap> >>> at Lorg/jikesrvm/runtime/Magic; >>> >>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >>> at line 166 >>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>> >>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/unboxed/Address; >>> at line 1003 >>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at line 61 >>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>> at Lorg/jikesrvm/classloader/RVMClass; >>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>> at Lorg/jikesrvm/jni/JNIFunctions; >>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/Address;)I >>> at line 185 >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> at <native frame> >>> JikesRVM: internal error: recursive use of hardware exception >>> registers (exiting) >>> >>> I also checked other Dacapo benches and they all were finished >>> correctly on this prototype configuration. This issue also reproduces >>> for me at r15100, r15000, r14950. The cattrack is down, so I don't >>> know whether this issue is reproduced on test machines. JIRA search >>> for "Dacapo chart" didn't bring me to any of relevant issues. >>> >>> Is this new or some known issue? >>> >>> Thanks, >>> Aleksey. >>> >>> ------------------------------------------------------------------------- >>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>> challenge >>> Build the coolest Linux based applications with Moblin SDK & win great >>> prizes >>> Grand prize is a trip for two to an Open Source event anywhere in the >>> world >>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>> _______________________________________________ >>> Jikesrvm-researchers mailing list >>> Jikesrvm-researchers@... >>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Jikesrvm-researchers mailing list >> Jikesrvm-researchers@... >> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >> > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Steve Blackburn
::
Rate this Message:
Hmmm....
Browsing the regressions mailing list, chart seems to be working fine in our nightly regressions (we're testing it each night and not failing on it) and I've also been testing it extensively in my own work. So my first guess is that it has something to do with Aleksey's environment :-/ Aleksey: have you noticed a regression, or has it never worked for you? I can try to take a closer look today. --Steve On 12/11/2008, at 1:25 AM, Aleksey Shipilev wrote: >> (I'm interested in jikesrvm.anu.edu.au)? > Argh, sorry, I meant jalapeno.anu.edu.au. > > On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev > <aleksey.shipilev@...> wrote: >> Thanks, Ian. >> >> Cattrack didn't catch any regressions on latest sanity check [1]. It >> looks like the problem is machine-dependent. Where can I get the info >> on exact HW/SW used on exact server (I'm interested in >> jikesrvm.anu.edu.au)? >> >> P.S. Cattrack seems to die at my hands... again. >> >> Thanks, >> Aleksey. >> >> [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart >> >> On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers >> <rogers.email@...> wrote: >>> Hi Aleksey, >>> >>> cattrack is back up again. I wasn't aware of any chart >>> regressions, but it >>> isn't executed as part of pre-commit due to its dependence on gtk- >>> peers in >>> Classpath (we don't build any peer sets on x86 64, on IA32 we do >>> build a >>> peer set iirc but it will only work if it sees a valid X server). >>> >>> Regards, >>> Ian >>> >>> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >>>> >>>> Hi, >>>> >>>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >>>> >>>> $ ant very-clean && bin/buildit --nuke localhost prototype >>>> >>>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and >>>> getting >>>> the failure: >>>> >>>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10- >>>> MR2.jar chart >>>> >>>> ===== DaCapo chart starting ===== >>>> >>>> Fatal error: NullPointerException within uninterruptible region. >>>> Exiting virtual machine due to uninterruptibility violation. >>>> >>>> -- Stack -- >>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>> deliverHardwareException(II)V at line 685 >>>> at <hardware trap> >>>> at Lorg/jikesrvm/runtime/Magic; >>>> >>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>> unboxed/Address; >>>> at line 166 >>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>> >>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>> unboxed/Address; >>>> at line 1003 >>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>> line 61 >>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>> at Lorg/jikesrvm/classloader/RVMClass; >>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>> Address;)I >>>> at line 185 >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> JikesRVM: internal error: recursive use of hardware exception >>>> registers (exiting) >>>> Scheduler.dumpStack(): in a recursive call, 2 deep. >>>> >>>> -- Stack -- >>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/ >>>> Address;)V >>>> at line 732 >>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >>>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at >>>> line >>>> 650 >>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>> traceback(Ljava/lang/String;)V at line 628 >>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>> deliverHardwareException(II)V at line 685 >>>> at <hardware trap> >>>> at Lorg/jikesrvm/runtime/Magic; >>>> >>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>> unboxed/Address; >>>> at line 166 >>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>> >>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>> unboxed/Address; >>>> at line 1003 >>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>> line 61 >>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>> at Lorg/jikesrvm/classloader/RVMClass; >>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>> Address;)I >>>> at line 185 >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> at <native frame> >>>> JikesRVM: internal error: recursive use of hardware exception >>>> registers (exiting) >>>> >>>> I also checked other Dacapo benches and they all were finished >>>> correctly on this prototype configuration. This issue also >>>> reproduces >>>> for me at r15100, r15000, r14950. The cattrack is down, so I don't >>>> know whether this issue is reproduced on test machines. JIRA search >>>> for "Dacapo chart" didn't bring me to any of relevant issues. >>>> >>>> Is this new or some known issue? >>>> >>>> Thanks, >>>> Aleksey. >>>> >>>> ------------------------------------------------------------------------- >>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>> challenge >>>> Build the coolest Linux based applications with Moblin SDK & win >>>> great >>>> prizes >>>> Grand prize is a trip for two to an Open Source event anywhere in >>>> the >>>> world >>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>> _______________________________________________ >>>> Jikesrvm-researchers mailing list >>>> Jikesrvm-researchers@... >>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>> >>> >>> ------------------------------------------------------------------------- >>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>> challenge >>> Build the coolest Linux based applications with Moblin SDK & win >>> great >>> prizes >>> Grand prize is a trip for two to an Open Source event anywhere in >>> the world >>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>> _______________________________________________ >>> Jikesrvm-researchers mailing list >>> Jikesrvm-researchers@... >>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>> >>> >> > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win > great prizes > Grand prize is a trip for two to an Open Source event anywhere in > the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Jikesrvm-researchers mailing list > Jikesrvm-researchers@... > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
Hi, Steve!
This is the first time I'm actually trying to run DaCapo:chart on JikesRVM+Classpath on my system. There were the problems with the headless, but I had migrated to machines where X servers are running well. What's interesting though, is that this kind of failure WAS observed when I reproduced DaCapo:chart failure on JikesRVM+Harmony on the same system [1]. It just flashed into my mind. I'm very positive that this one is related to my environment. I would appreciate any guidance on getting this bug fixed -- so far I have very inconclusive failure on stack enumeration (?) connected with class loading (?). Now I'm trying to grasp what class is being looked up via FindClass(...). I can supply anyone with precise HW/SW environment if it would help (actually I doubt that). Another option is, I can provide SSH to anyone still interested in the issue. Of course, if I would fail first ;) Thanks, Aleksey. [1] http://jira.codehaus.org/browse/RVM-618?focusedCommentId=144524#action_144524 On Tue, Nov 11, 2008 at 10:51 PM, Steve Blackburn <Steve.Blackburn@...> wrote: > Hmmm.... > > Browsing the regressions mailing list, chart seems to be working fine > in our nightly regressions (we're testing it each night and not > failing on it) and I've also been testing it extensively in my own > work. So my first guess is that it has something to do with > Aleksey's environment :-/ > > Aleksey: have you noticed a regression, or has it never worked for you? > > I can try to take a closer look today. > > --Steve > > On 12/11/2008, at 1:25 AM, Aleksey Shipilev wrote: > >>> (I'm interested in jikesrvm.anu.edu.au)? >> Argh, sorry, I meant jalapeno.anu.edu.au. >> >> On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev >> <aleksey.shipilev@...> wrote: >>> Thanks, Ian. >>> >>> Cattrack didn't catch any regressions on latest sanity check [1]. It >>> looks like the problem is machine-dependent. Where can I get the info >>> on exact HW/SW used on exact server (I'm interested in >>> jikesrvm.anu.edu.au)? >>> >>> P.S. Cattrack seems to die at my hands... again. >>> >>> Thanks, >>> Aleksey. >>> >>> [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart >>> >>> On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers >>> <rogers.email@...> wrote: >>>> Hi Aleksey, >>>> >>>> cattrack is back up again. I wasn't aware of any chart >>>> regressions, but it >>>> isn't executed as part of pre-commit due to its dependence on gtk- >>>> peers in >>>> Classpath (we don't build any peer sets on x86 64, on IA32 we do >>>> build a >>>> peer set iirc but it will only work if it sees a valid X server). >>>> >>>> Regards, >>>> Ian >>>> >>>> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >>>>> >>>>> Hi, >>>>> >>>>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >>>>> >>>>> $ ant very-clean && bin/buildit --nuke localhost prototype >>>>> >>>>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and >>>>> getting >>>>> the failure: >>>>> >>>>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10- >>>>> MR2.jar chart >>>>> >>>>> ===== DaCapo chart starting ===== >>>>> >>>>> Fatal error: NullPointerException within uninterruptible region. >>>>> Exiting virtual machine due to uninterruptibility violation. >>>>> >>>>> -- Stack -- >>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>> deliverHardwareException(II)V at line 685 >>>>> at <hardware trap> >>>>> at Lorg/jikesrvm/runtime/Magic; >>>>> >>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>> unboxed/Address; >>>>> at line 166 >>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>> >>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>> unboxed/Address; >>>>> at line 1003 >>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>> line 61 >>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>> Address;)I >>>>> at line 185 >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> JikesRVM: internal error: recursive use of hardware exception >>>>> registers (exiting) >>>>> Scheduler.dumpStack(): in a recursive call, 2 deep. >>>>> >>>>> -- Stack -- >>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/ >>>>> Address;)V >>>>> at line 732 >>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >>>>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at >>>>> line >>>>> 650 >>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>> traceback(Ljava/lang/String;)V at line 628 >>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>> deliverHardwareException(II)V at line 685 >>>>> at <hardware trap> >>>>> at Lorg/jikesrvm/runtime/Magic; >>>>> >>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>> unboxed/Address; >>>>> at line 166 >>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>> >>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>> unboxed/Address; >>>>> at line 1003 >>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>> line 61 >>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>> Address;)I >>>>> at line 185 >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> at <native frame> >>>>> JikesRVM: internal error: recursive use of hardware exception >>>>> registers (exiting) >>>>> >>>>> I also checked other Dacapo benches and they all were finished >>>>> correctly on this prototype configuration. This issue also >>>>> reproduces >>>>> for me at r15100, r15000, r14950. The cattrack is down, so I don't >>>>> know whether this issue is reproduced on test machines. JIRA search >>>>> for "Dacapo chart" didn't bring me to any of relevant issues. >>>>> >>>>> Is this new or some known issue? >>>>> >>>>> Thanks, >>>>> Aleksey. >>>>> >>>>> ------------------------------------------------------------------------- >>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>> challenge >>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>> great >>>>> prizes >>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>> the >>>>> world >>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>> _______________________________________________ >>>>> Jikesrvm-researchers mailing list >>>>> Jikesrvm-researchers@... >>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>> >>>> >>>> ------------------------------------------------------------------------- >>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>> challenge >>>> Build the coolest Linux based applications with Moblin SDK & win >>>> great >>>> prizes >>>> Grand prize is a trip for two to an Open Source event anywhere in >>>> the world >>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>> _______________________________________________ >>>> Jikesrvm-researchers mailing list >>>> Jikesrvm-researchers@... >>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>> >>>> >>> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win >> great prizes >> Grand prize is a trip for two to an Open Source event anywhere in >> the world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Jikesrvm-researchers mailing list >> Jikesrvm-researchers@... >> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Jikesrvm-researchers mailing list > Jikesrvm-researchers@... > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
Aha, running with -verbose:jni:
===== DaCapo chart starting ===== [Dynamic-linking native method gnu.java.nio.VMChannel.available (I)I [Loaded native library: /home/shade/trunks/jikesrvm/dist/prototype_ia32-linux/libgtkpeer.so] [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkToolkit.initIDs ()V [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkToolkit.gtkInit (ILjava/lang/Object;)V [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkGenericPeer.initIDs ()V Fatal error: NullPointerException within uninterruptible region. Exiting virtual machine due to uninterruptibility violation. *CRASH* Should I do something with GNU Classpath on that matter? DaCapo page [1] says I need to enable gtk-peer to run chart -- and judging on verbose output, it is. Thanks, Aleksey. [1] http://dacapobench.org/problems.html On Tue, Nov 11, 2008 at 11:48 PM, Aleksey Shipilev <aleksey.shipilev@...> wrote: > Hi, Steve! > > This is the first time I'm actually trying to run DaCapo:chart on > JikesRVM+Classpath on my system. There were the problems with the > headless, but I had migrated to machines where X servers are running > well. What's interesting though, is that this kind of failure WAS > observed when I reproduced DaCapo:chart failure on JikesRVM+Harmony on > the same system [1]. It just flashed into my mind. > > I'm very positive that this one is related to my environment. I would > appreciate any guidance on getting this bug fixed -- so far I have > very inconclusive failure on stack enumeration (?) connected with > class loading (?). Now I'm trying to grasp what class is being looked > up via FindClass(...). > > I can supply anyone with precise HW/SW environment if it would help > (actually I doubt that). Another option is, I can provide SSH to > anyone still interested in the issue. Of course, if I would fail first > ;) > > Thanks, > Aleksey. > > [1] http://jira.codehaus.org/browse/RVM-618?focusedCommentId=144524#action_144524 > > On Tue, Nov 11, 2008 at 10:51 PM, Steve Blackburn > <Steve.Blackburn@...> wrote: >> Hmmm.... >> >> Browsing the regressions mailing list, chart seems to be working fine >> in our nightly regressions (we're testing it each night and not >> failing on it) and I've also been testing it extensively in my own >> work. So my first guess is that it has something to do with >> Aleksey's environment :-/ >> >> Aleksey: have you noticed a regression, or has it never worked for you? >> >> I can try to take a closer look today. >> >> --Steve >> >> On 12/11/2008, at 1:25 AM, Aleksey Shipilev wrote: >> >>>> (I'm interested in jikesrvm.anu.edu.au)? >>> Argh, sorry, I meant jalapeno.anu.edu.au. >>> >>> On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev >>> <aleksey.shipilev@...> wrote: >>>> Thanks, Ian. >>>> >>>> Cattrack didn't catch any regressions on latest sanity check [1]. It >>>> looks like the problem is machine-dependent. Where can I get the info >>>> on exact HW/SW used on exact server (I'm interested in >>>> jikesrvm.anu.edu.au)? >>>> >>>> P.S. Cattrack seems to die at my hands... again. >>>> >>>> Thanks, >>>> Aleksey. >>>> >>>> [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart >>>> >>>> On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers >>>> <rogers.email@...> wrote: >>>>> Hi Aleksey, >>>>> >>>>> cattrack is back up again. I wasn't aware of any chart >>>>> regressions, but it >>>>> isn't executed as part of pre-commit due to its dependence on gtk- >>>>> peers in >>>>> Classpath (we don't build any peer sets on x86 64, on IA32 we do >>>>> build a >>>>> peer set iirc but it will only work if it sees a valid X server). >>>>> >>>>> Regards, >>>>> Ian >>>>> >>>>> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >>>>>> >>>>>> Hi, >>>>>> >>>>>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >>>>>> >>>>>> $ ant very-clean && bin/buildit --nuke localhost prototype >>>>>> >>>>>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and >>>>>> getting >>>>>> the failure: >>>>>> >>>>>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10- >>>>>> MR2.jar chart >>>>>> >>>>>> ===== DaCapo chart starting ===== >>>>>> >>>>>> Fatal error: NullPointerException within uninterruptible region. >>>>>> Exiting virtual machine due to uninterruptibility violation. >>>>>> >>>>>> -- Stack -- >>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>> deliverHardwareException(II)V at line 685 >>>>>> at <hardware trap> >>>>>> at Lorg/jikesrvm/runtime/Magic; >>>>>> >>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>> unboxed/Address; >>>>>> at line 166 >>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>> >>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>> unboxed/Address; >>>>>> at line 1003 >>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>>> line 61 >>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>>> Address;)I >>>>>> at line 185 >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> JikesRVM: internal error: recursive use of hardware exception >>>>>> registers (exiting) >>>>>> Scheduler.dumpStack(): in a recursive call, 2 deep. >>>>>> >>>>>> -- Stack -- >>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/ >>>>>> Address;)V >>>>>> at line 732 >>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >>>>>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at >>>>>> line >>>>>> 650 >>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>> traceback(Ljava/lang/String;)V at line 628 >>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>> deliverHardwareException(II)V at line 685 >>>>>> at <hardware trap> >>>>>> at Lorg/jikesrvm/runtime/Magic; >>>>>> >>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>> unboxed/Address; >>>>>> at line 166 >>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>> >>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>> unboxed/Address; >>>>>> at line 1003 >>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>>> line 61 >>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>>> Address;)I >>>>>> at line 185 >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> at <native frame> >>>>>> JikesRVM: internal error: recursive use of hardware exception >>>>>> registers (exiting) >>>>>> >>>>>> I also checked other Dacapo benches and they all were finished >>>>>> correctly on this prototype configuration. This issue also >>>>>> reproduces >>>>>> for me at r15100, r15000, r14950. The cattrack is down, so I don't >>>>>> know whether this issue is reproduced on test machines. JIRA search >>>>>> for "Dacapo chart" didn't bring me to any of relevant issues. >>>>>> >>>>>> Is this new or some known issue? >>>>>> >>>>>> Thanks, >>>>>> Aleksey. >>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>>> challenge >>>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>>> great >>>>>> prizes >>>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>>> the >>>>>> world >>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>>> _______________________________________________ >>>>>> Jikesrvm-researchers mailing list >>>>>> Jikesrvm-researchers@... >>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>>> >>>>> >>>>> ------------------------------------------------------------------------- >>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>> challenge >>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>> great >>>>> prizes >>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>> the world >>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>> _______________________________________________ >>>>> Jikesrvm-researchers mailing list >>>>> Jikesrvm-researchers@... >>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>>> >>>>> >>>> >>> >>> ------------------------------------------------------------------------- >>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>> challenge >>> Build the coolest Linux based applications with Moblin SDK & win >>> great prizes >>> Grand prize is a trip for two to an Open Source event anywhere in >>> the world >>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>> _______________________________________________ >>> Jikesrvm-researchers mailing list >>> Jikesrvm-researchers@... >>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >> Build the coolest Linux based applications with Moblin SDK & win great prizes >> Grand prize is a trip for two to an Open Source event anywhere in the world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Jikesrvm-researchers mailing list >> Jikesrvm-researchers@... >> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
So we have a pleasant remote debugging session with Ian (thanks!).
The instrumentation of org.jikesrvm.runtime.Magic.getReturnAddress(Address fp) gets us to the point where fp == 0 right before the NPE. Ian assumes that's the issue with lack of frame pointers and the way RVM walks the stack. That idea is supported by the fact in my environment GTK and related packages are built with -fomit-frame-pointer. I'm now rebuilding my dev-system with frame pointers for all relevant packages to test that claim. I will update here with the results. Thanks, Aleksey. On Wed, Nov 12, 2008 at 12:10 AM, Aleksey Shipilev <aleksey.shipilev@...> wrote: > Aha, running with -verbose:jni: > > ===== DaCapo chart starting ===== > [Dynamic-linking native method gnu.java.nio.VMChannel.available (I)I > [Loaded native library: > /home/shade/trunks/jikesrvm/dist/prototype_ia32-linux/libgtkpeer.so] > [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkToolkit.initIDs ()V > [Dynamic-linking native method > gnu.java.awt.peer.gtk.GtkToolkit.gtkInit (ILjava/lang/Object;)V > [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkGenericPeer.initIDs ()V > > Fatal error: NullPointerException within uninterruptible region. > Exiting virtual machine due to uninterruptibility violation. > *CRASH* > > > Should I do something with GNU Classpath on that matter? > DaCapo page [1] says I need to enable gtk-peer to run chart -- and > judging on verbose output, it is. > > Thanks, > Aleksey. > > [1] http://dacapobench.org/problems.html > > On Tue, Nov 11, 2008 at 11:48 PM, Aleksey Shipilev > <aleksey.shipilev@...> wrote: >> Hi, Steve! >> >> This is the first time I'm actually trying to run DaCapo:chart on >> JikesRVM+Classpath on my system. There were the problems with the >> headless, but I had migrated to machines where X servers are running >> well. What's interesting though, is that this kind of failure WAS >> observed when I reproduced DaCapo:chart failure on JikesRVM+Harmony on >> the same system [1]. It just flashed into my mind. >> >> I'm very positive that this one is related to my environment. I would >> appreciate any guidance on getting this bug fixed -- so far I have >> very inconclusive failure on stack enumeration (?) connected with >> class loading (?). Now I'm trying to grasp what class is being looked >> up via FindClass(...). >> >> I can supply anyone with precise HW/SW environment if it would help >> (actually I doubt that). Another option is, I can provide SSH to >> anyone still interested in the issue. Of course, if I would fail first >> ;) >> >> Thanks, >> Aleksey. >> >> [1] http://jira.codehaus.org/browse/RVM-618?focusedCommentId=144524#action_144524 >> >> On Tue, Nov 11, 2008 at 10:51 PM, Steve Blackburn >> <Steve.Blackburn@...> wrote: >>> Hmmm.... >>> >>> Browsing the regressions mailing list, chart seems to be working fine >>> in our nightly regressions (we're testing it each night and not >>> failing on it) and I've also been testing it extensively in my own >>> work. So my first guess is that it has something to do with >>> Aleksey's environment :-/ >>> >>> Aleksey: have you noticed a regression, or has it never worked for you? >>> >>> I can try to take a closer look today. >>> >>> --Steve >>> >>> On 12/11/2008, at 1:25 AM, Aleksey Shipilev wrote: >>> >>>>> (I'm interested in jikesrvm.anu.edu.au)? >>>> Argh, sorry, I meant jalapeno.anu.edu.au. >>>> >>>> On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev >>>> <aleksey.shipilev@...> wrote: >>>>> Thanks, Ian. >>>>> >>>>> Cattrack didn't catch any regressions on latest sanity check [1]. It >>>>> looks like the problem is machine-dependent. Where can I get the info >>>>> on exact HW/SW used on exact server (I'm interested in >>>>> jikesrvm.anu.edu.au)? >>>>> >>>>> P.S. Cattrack seems to die at my hands... again. >>>>> >>>>> Thanks, >>>>> Aleksey. >>>>> >>>>> [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart >>>>> >>>>> On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers >>>>> <rogers.email@...> wrote: >>>>>> Hi Aleksey, >>>>>> >>>>>> cattrack is back up again. I wasn't aware of any chart >>>>>> regressions, but it >>>>>> isn't executed as part of pre-commit due to its dependence on gtk- >>>>>> peers in >>>>>> Classpath (we don't build any peer sets on x86 64, on IA32 we do >>>>>> build a >>>>>> peer set iirc but it will only work if it sees a valid X server). >>>>>> >>>>>> Regards, >>>>>> Ian >>>>>> >>>>>> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >>>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >>>>>>> >>>>>>> $ ant very-clean && bin/buildit --nuke localhost prototype >>>>>>> >>>>>>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and >>>>>>> getting >>>>>>> the failure: >>>>>>> >>>>>>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10- >>>>>>> MR2.jar chart >>>>>>> >>>>>>> ===== DaCapo chart starting ===== >>>>>>> >>>>>>> Fatal error: NullPointerException within uninterruptible region. >>>>>>> Exiting virtual machine due to uninterruptibility violation. >>>>>>> >>>>>>> -- Stack -- >>>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>> deliverHardwareException(II)V at line 685 >>>>>>> at <hardware trap> >>>>>>> at Lorg/jikesrvm/runtime/Magic; >>>>>>> >>>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>> unboxed/Address; >>>>>>> at line 166 >>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>> >>>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>> unboxed/Address; >>>>>>> at line 1003 >>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>>>> line 61 >>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>>>> Address;)I >>>>>>> at line 185 >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> JikesRVM: internal error: recursive use of hardware exception >>>>>>> registers (exiting) >>>>>>> Scheduler.dumpStack(): in a recursive call, 2 deep. >>>>>>> >>>>>>> -- Stack -- >>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/ >>>>>>> Address;)V >>>>>>> at line 732 >>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at >>>>>>> line >>>>>>> 650 >>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>>> traceback(Ljava/lang/String;)V at line 628 >>>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>> deliverHardwareException(II)V at line 685 >>>>>>> at <hardware trap> >>>>>>> at Lorg/jikesrvm/runtime/Magic; >>>>>>> >>>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>> unboxed/Address; >>>>>>> at line 166 >>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>> >>>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>> unboxed/Address; >>>>>>> at line 1003 >>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>>>> line 61 >>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>>>> Address;)I >>>>>>> at line 185 >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> at <native frame> >>>>>>> JikesRVM: internal error: recursive use of hardware exception >>>>>>> registers (exiting) >>>>>>> >>>>>>> I also checked other Dacapo benches and they all were finished >>>>>>> correctly on this prototype configuration. This issue also >>>>>>> reproduces >>>>>>> for me at r15100, r15000, r14950. The cattrack is down, so I don't >>>>>>> know whether this issue is reproduced on test machines. JIRA search >>>>>>> for "Dacapo chart" didn't bring me to any of relevant issues. >>>>>>> >>>>>>> Is this new or some known issue? >>>>>>> >>>>>>> Thanks, >>>>>>> Aleksey. >>>>>>> >>>>>>> ------------------------------------------------------------------------- >>>>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>>>> challenge >>>>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>>>> great >>>>>>> prizes >>>>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>>>> the >>>>>>> world >>>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>>>> _______________________________________________ >>>>>>> Jikesrvm-researchers mailing list >>>>>>> Jikesrvm-researchers@... >>>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>>>> >>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>>> challenge >>>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>>> great >>>>>> prizes >>>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>>> the world >>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>>> _______________________________________________ >>>>>> Jikesrvm-researchers mailing list >>>>>> Jikesrvm-researchers@... >>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>>>> >>>>>> >>>>> >>>> >>>> ------------------------------------------------------------------------- >>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>> challenge >>>> Build the coolest Linux based applications with Moblin SDK & win >>>> great prizes >>>> Grand prize is a trip for two to an Open Source event anywhere in >>>> the world >>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>> _______________________________________________ >>>> Jikesrvm-researchers mailing list >>>> Jikesrvm-researchers@... >>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>> >>> >>> ------------------------------------------------------------------------- >>> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >>> Build the coolest Linux based applications with Moblin SDK & win great prizes >>> Grand prize is a trip for two to an Open Source event anywhere in the world >>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>> _______________________________________________ >>> Jikesrvm-researchers mailing list >>> Jikesrvm-researchers@... >>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>> >> > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
I had rebuilt packages on which GNU classpath relies (GTK+, GL, part
of X server, whatever Gentoo finds as dependent), and DaCapo:chart runs well! AFAIK, Ian now tries to implement stack walking without frame-pointers, that would be clearer and wiser things to do. I assume the issue is closed for me, but I want to express this knowledge somewhere, before bug is fixed. Is there some place on wiki for this? Or just some JIRA will be enough? Thanks, Aleksey. On Wed, Nov 12, 2008 at 1:30 AM, Aleksey Shipilev <aleksey.shipilev@...> wrote: > So we have a pleasant remote debugging session with Ian (thanks!). > > The instrumentation of > org.jikesrvm.runtime.Magic.getReturnAddress(Address fp) gets us to the > point where fp == 0 right before the NPE. Ian assumes that's the issue > with lack of frame pointers and the way RVM walks the stack. That idea > is supported by the fact in my environment GTK and related packages > are built with -fomit-frame-pointer. I'm now rebuilding my dev-system > with frame pointers for all relevant packages to test that claim. > > I will update here with the results. > > Thanks, > Aleksey. > > On Wed, Nov 12, 2008 at 12:10 AM, Aleksey Shipilev > <aleksey.shipilev@...> wrote: >> Aha, running with -verbose:jni: >> >> ===== DaCapo chart starting ===== >> [Dynamic-linking native method gnu.java.nio.VMChannel.available (I)I >> [Loaded native library: >> /home/shade/trunks/jikesrvm/dist/prototype_ia32-linux/libgtkpeer.so] >> [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkToolkit.initIDs ()V >> [Dynamic-linking native method >> gnu.java.awt.peer.gtk.GtkToolkit.gtkInit (ILjava/lang/Object;)V >> [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkGenericPeer.initIDs ()V >> >> Fatal error: NullPointerException within uninterruptible region. >> Exiting virtual machine due to uninterruptibility violation. >> *CRASH* >> >> >> Should I do something with GNU Classpath on that matter? >> DaCapo page [1] says I need to enable gtk-peer to run chart -- and >> judging on verbose output, it is. >> >> Thanks, >> Aleksey. >> >> [1] http://dacapobench.org/problems.html >> >> On Tue, Nov 11, 2008 at 11:48 PM, Aleksey Shipilev >> <aleksey.shipilev@...> wrote: >>> Hi, Steve! >>> >>> This is the first time I'm actually trying to run DaCapo:chart on >>> JikesRVM+Classpath on my system. There were the problems with the >>> headless, but I had migrated to machines where X servers are running >>> well. What's interesting though, is that this kind of failure WAS >>> observed when I reproduced DaCapo:chart failure on JikesRVM+Harmony on >>> the same system [1]. It just flashed into my mind. >>> >>> I'm very positive that this one is related to my environment. I would >>> appreciate any guidance on getting this bug fixed -- so far I have >>> very inconclusive failure on stack enumeration (?) connected with >>> class loading (?). Now I'm trying to grasp what class is being looked >>> up via FindClass(...). >>> >>> I can supply anyone with precise HW/SW environment if it would help >>> (actually I doubt that). Another option is, I can provide SSH to >>> anyone still interested in the issue. Of course, if I would fail first >>> ;) >>> >>> Thanks, >>> Aleksey. >>> >>> [1] http://jira.codehaus.org/browse/RVM-618?focusedCommentId=144524#action_144524 >>> >>> On Tue, Nov 11, 2008 at 10:51 PM, Steve Blackburn >>> <Steve.Blackburn@...> wrote: >>>> Hmmm.... >>>> >>>> Browsing the regressions mailing list, chart seems to be working fine >>>> in our nightly regressions (we're testing it each night and not >>>> failing on it) and I've also been testing it extensively in my own >>>> work. So my first guess is that it has something to do with >>>> Aleksey's environment :-/ >>>> >>>> Aleksey: have you noticed a regression, or has it never worked for you? >>>> >>>> I can try to take a closer look today. >>>> >>>> --Steve >>>> >>>> On 12/11/2008, at 1:25 AM, Aleksey Shipilev wrote: >>>> >>>>>> (I'm interested in jikesrvm.anu.edu.au)? >>>>> Argh, sorry, I meant jalapeno.anu.edu.au. >>>>> >>>>> On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev >>>>> <aleksey.shipilev@...> wrote: >>>>>> Thanks, Ian. >>>>>> >>>>>> Cattrack didn't catch any regressions on latest sanity check [1]. It >>>>>> looks like the problem is machine-dependent. Where can I get the info >>>>>> on exact HW/SW used on exact server (I'm interested in >>>>>> jikesrvm.anu.edu.au)? >>>>>> >>>>>> P.S. Cattrack seems to die at my hands... again. >>>>>> >>>>>> Thanks, >>>>>> Aleksey. >>>>>> >>>>>> [1] http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart >>>>>> >>>>>> On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers >>>>>> <rogers.email@...> wrote: >>>>>>> Hi Aleksey, >>>>>>> >>>>>>> cattrack is back up again. I wasn't aware of any chart >>>>>>> regressions, but it >>>>>>> isn't executed as part of pre-commit due to its dependence on gtk- >>>>>>> peers in >>>>>>> Classpath (we don't build any peer sets on x86 64, on IA32 we do >>>>>>> build a >>>>>>> peer set iirc but it will only work if it sees a valid X server). >>>>>>> >>>>>>> Regards, >>>>>>> Ian >>>>>>> >>>>>>> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >>>>>>>> >>>>>>>> Hi, >>>>>>>> >>>>>>>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >>>>>>>> >>>>>>>> $ ant very-clean && bin/buildit --nuke localhost prototype >>>>>>>> >>>>>>>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and >>>>>>>> getting >>>>>>>> the failure: >>>>>>>> >>>>>>>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10- >>>>>>>> MR2.jar chart >>>>>>>> >>>>>>>> ===== DaCapo chart starting ===== >>>>>>>> >>>>>>>> Fatal error: NullPointerException within uninterruptible region. >>>>>>>> Exiting virtual machine due to uninterruptibility violation. >>>>>>>> >>>>>>>> -- Stack -- >>>>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>>> deliverHardwareException(II)V at line 685 >>>>>>>> at <hardware trap> >>>>>>>> at Lorg/jikesrvm/runtime/Magic; >>>>>>>> >>>>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>>> unboxed/Address; >>>>>>>> at line 166 >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>>> >>>>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>>> unboxed/Address; >>>>>>>> at line 1003 >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>>>>> line 61 >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>>>>> Address;)I >>>>>>>> at line 185 >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> JikesRVM: internal error: recursive use of hardware exception >>>>>>>> registers (exiting) >>>>>>>> Scheduler.dumpStack(): in a recursive call, 2 deep. >>>>>>>> >>>>>>>> -- Stack -- >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/ >>>>>>>> Address;)V >>>>>>>> at line 732 >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at >>>>>>>> line >>>>>>>> 650 >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >>>>>>>> traceback(Ljava/lang/String;)V at line 628 >>>>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>>> deliverHardwareException(II)V at line 685 >>>>>>>> at <hardware trap> >>>>>>>> at Lorg/jikesrvm/runtime/Magic; >>>>>>>> >>>>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>>> unboxed/Address; >>>>>>>> at line 166 >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >>>>>>>> >>>>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >>>>>>>> unboxed/Address; >>>>>>>> at line 1003 >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >>>>>>>> line 61 >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >>>>>>>> at Lorg/jikesrvm/classloader/RVMClass; >>>>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line 1119 >>>>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >>>>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >>>>>>>> Address;)I >>>>>>>> at line 185 >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> at <native frame> >>>>>>>> JikesRVM: internal error: recursive use of hardware exception >>>>>>>> registers (exiting) >>>>>>>> >>>>>>>> I also checked other Dacapo benches and they all were finished >>>>>>>> correctly on this prototype configuration. This issue also >>>>>>>> reproduces >>>>>>>> for me at r15100, r15000, r14950. The cattrack is down, so I don't >>>>>>>> know whether this issue is reproduced on test machines. JIRA search >>>>>>>> for "Dacapo chart" didn't bring me to any of relevant issues. >>>>>>>> >>>>>>>> Is this new or some known issue? >>>>>>>> >>>>>>>> Thanks, >>>>>>>> Aleksey. >>>>>>>> >>>>>>>> ------------------------------------------------------------------------- >>>>>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>>>>> challenge >>>>>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>>>>> great >>>>>>>> prizes >>>>>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>>>>> the >>>>>>>> world >>>>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>>>>> _______________________________________________ >>>>>>>> Jikesrvm-researchers mailing list >>>>>>>> Jikesrvm-researchers@... >>>>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>>>>> >>>>>>> >>>>>>> ------------------------------------------------------------------------- >>>>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>>>> challenge >>>>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>>>> great >>>>>>> prizes >>>>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>>>> the world >>>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>>>> _______________________________________________ >>>>>>> Jikesrvm-researchers mailing list >>>>>>> Jikesrvm-researchers@... >>>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>>>>> >>>>>>> >>>>>> >>>>> >>>>> ------------------------------------------------------------------------- >>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>>>> challenge >>>>> Build the coolest Linux based applications with Moblin SDK & win >>>>> great prizes >>>>> Grand prize is a trip for two to an Open Source event anywhere in >>>>> the world >>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>>> _______________________________________________ >>>>> Jikesrvm-researchers mailing list >>>>> Jikesrvm-researchers@... >>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>> >>>> >>>> ------------------------------------------------------------------------- >>>> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >>>> Build the coolest Linux based applications with Moblin SDK & win great prizes >>>> Grand prize is a trip for two to an Open Source event anywhere in the world >>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>>> _______________________________________________ >>>> Jikesrvm-researchers mailing list >>>> Jikesrvm-researchers@... >>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >>>> >>> >> > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Ian Rogers
::
Rate this Message:
Hi Aleksey,
I guess this is a build requirement so you could make some notes in the wiki. I haven't yet got the new frame pointer code into the trunk. Thanks for your perseverance with this and helping us isolate this as not a bug in the strictest of senses. Regards, Ian 2008/11/12 Aleksey Shipilev <aleksey.shipilev@...> I had rebuilt packages on which GNU classpath relies (GTK+, GL, part ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
|
|
Re: [rvm-research] DaCapo:chart is broken?
by Aleksey Shipilev
::
Rate this Message:
Yeah, that's not the bug in general sense.
Just from frustration standpoint is more like buggy thing :) I had updated "Building RVM": http://docs.codehaus.org/pages/diffpages.action?pageId=71513&originalId=111182972 Thanks, Aleksey. On Wed, Nov 12, 2008 at 1:27 PM, Ian Rogers <rogers.email@...> wrote: > Hi Aleksey, > > I guess this is a build requirement so you could make some notes in the > wiki. I haven't yet got the new frame pointer code into the trunk. Thanks > for your perseverance with this and helping us isolate this as not a bug in > the strictest of senses. > > Regards, > Ian > > 2008/11/12 Aleksey Shipilev <aleksey.shipilev@...> >> >> I had rebuilt packages on which GNU classpath relies (GTK+, GL, part >> of X server, whatever Gentoo finds as dependent), and DaCapo:chart >> runs well! >> >> AFAIK, Ian now tries to implement stack walking without >> frame-pointers, that would be clearer and wiser things to do. I assume >> the issue is closed for me, but I want to express this knowledge >> somewhere, before bug is fixed. Is there some place on wiki for this? >> Or just some JIRA will be enough? >> >> Thanks, >> Aleksey. >> >> On Wed, Nov 12, 2008 at 1:30 AM, Aleksey Shipilev >> <aleksey.shipilev@...> wrote: >> > So we have a pleasant remote debugging session with Ian (thanks!). >> > >> > The instrumentation of >> > org.jikesrvm.runtime.Magic.getReturnAddress(Address fp) gets us to the >> > point where fp == 0 right before the NPE. Ian assumes that's the issue >> > with lack of frame pointers and the way RVM walks the stack. That idea >> > is supported by the fact in my environment GTK and related packages >> > are built with -fomit-frame-pointer. I'm now rebuilding my dev-system >> > with frame pointers for all relevant packages to test that claim. >> > >> > I will update here with the results. >> > >> > Thanks, >> > Aleksey. >> > >> > On Wed, Nov 12, 2008 at 12:10 AM, Aleksey Shipilev >> > <aleksey.shipilev@...> wrote: >> >> Aha, running with -verbose:jni: >> >> >> >> ===== DaCapo chart starting ===== >> >> [Dynamic-linking native method gnu.java.nio.VMChannel.available (I)I >> >> [Loaded native library: >> >> /home/shade/trunks/jikesrvm/dist/prototype_ia32-linux/libgtkpeer.so] >> >> [Dynamic-linking native method gnu.java.awt.peer.gtk.GtkToolkit.initIDs >> >> ()V >> >> [Dynamic-linking native method >> >> gnu.java.awt.peer.gtk.GtkToolkit.gtkInit (ILjava/lang/Object;)V >> >> [Dynamic-linking native method >> >> gnu.java.awt.peer.gtk.GtkGenericPeer.initIDs ()V >> >> >> >> Fatal error: NullPointerException within uninterruptible region. >> >> Exiting virtual machine due to uninterruptibility violation. >> >> *CRASH* >> >> >> >> >> >> Should I do something with GNU Classpath on that matter? >> >> DaCapo page [1] says I need to enable gtk-peer to run chart -- and >> >> judging on verbose output, it is. >> >> >> >> Thanks, >> >> Aleksey. >> >> >> >> [1] http://dacapobench.org/problems.html >> >> >> >> On Tue, Nov 11, 2008 at 11:48 PM, Aleksey Shipilev >> >> <aleksey.shipilev@...> wrote: >> >>> Hi, Steve! >> >>> >> >>> This is the first time I'm actually trying to run DaCapo:chart on >> >>> JikesRVM+Classpath on my system. There were the problems with the >> >>> headless, but I had migrated to machines where X servers are running >> >>> well. What's interesting though, is that this kind of failure WAS >> >>> observed when I reproduced DaCapo:chart failure on JikesRVM+Harmony on >> >>> the same system [1]. It just flashed into my mind. >> >>> >> >>> I'm very positive that this one is related to my environment. I would >> >>> appreciate any guidance on getting this bug fixed -- so far I have >> >>> very inconclusive failure on stack enumeration (?) connected with >> >>> class loading (?). Now I'm trying to grasp what class is being looked >> >>> up via FindClass(...). >> >>> >> >>> I can supply anyone with precise HW/SW environment if it would help >> >>> (actually I doubt that). Another option is, I can provide SSH to >> >>> anyone still interested in the issue. Of course, if I would fail first >> >>> ;) >> >>> >> >>> Thanks, >> >>> Aleksey. >> >>> >> >>> [1] >> >>> http://jira.codehaus.org/browse/RVM-618?focusedCommentId=144524#action_144524 >> >>> >> >>> On Tue, Nov 11, 2008 at 10:51 PM, Steve Blackburn >> >>> <Steve.Blackburn@...> wrote: >> >>>> Hmmm.... >> >>>> >> >>>> Browsing the regressions mailing list, chart seems to be working fine >> >>>> in our nightly regressions (we're testing it each night and not >> >>>> failing on it) and I've also been testing it extensively in my own >> >>>> work. So my first guess is that it has something to do with >> >>>> Aleksey's environment :-/ >> >>>> >> >>>> Aleksey: have you noticed a regression, or has it never worked for >> >>>> you? >> >>>> >> >>>> I can try to take a closer look today. >> >>>> >> >>>> --Steve >> >>>> >> >>>> On 12/11/2008, at 1:25 AM, Aleksey Shipilev wrote: >> >>>> >> >>>>>> (I'm interested in jikesrvm.anu.edu.au)? >> >>>>> Argh, sorry, I meant jalapeno.anu.edu.au. >> >>>>> >> >>>>> On Tue, Nov 11, 2008 at 5:21 PM, Aleksey Shipilev >> >>>>> <aleksey.shipilev@...> wrote: >> >>>>>> Thanks, Ian. >> >>>>>> >> >>>>>> Cattrack didn't catch any regressions on latest sanity check [1]. >> >>>>>> It >> >>>>>> looks like the problem is machine-dependent. Where can I get the >> >>>>>> info >> >>>>>> on exact HW/SW used on exact server (I'm interested in >> >>>>>> jikesrvm.anu.edu.au)? >> >>>>>> >> >>>>>> P.S. Cattrack seems to die at my hands... again. >> >>>>>> >> >>>>>> Thanks, >> >>>>>> Aleksey. >> >>>>>> >> >>>>>> [1] >> >>>>>> http://jikesrvm.anu.edu.au/cattrack/results/jalapeno.anu.edu.au/sanity/6243/prototype/default/dacapo/chart >> >>>>>> >> >>>>>> On Tue, Nov 11, 2008 at 4:28 PM, Ian Rogers >> >>>>>> <rogers.email@...> wrote: >> >>>>>>> Hi Aleksey, >> >>>>>>> >> >>>>>>> cattrack is back up again. I wasn't aware of any chart >> >>>>>>> regressions, but it >> >>>>>>> isn't executed as part of pre-commit due to its dependence on gtk- >> >>>>>>> peers in >> >>>>>>> Classpath (we don't build any peer sets on x86 64, on IA32 we do >> >>>>>>> build a >> >>>>>>> peer set iirc but it will only work if it sees a valid X server). >> >>>>>>> >> >>>>>>> Regards, >> >>>>>>> Ian >> >>>>>>> >> >>>>>>> 2008/11/11 Aleksey Shipilev <aleksey.shipilev@...> >> >>>>>>>> >> >>>>>>>> Hi, >> >>>>>>>> >> >>>>>>>> I had checked out JikesRVM r15150 (HEAD today) and built it with: >> >>>>>>>> >> >>>>>>>> $ ant very-clean && bin/buildit --nuke localhost prototype >> >>>>>>>> >> >>>>>>>> after that, I'm trying to run DaCapo:chart on Gentoo x86, and >> >>>>>>>> getting >> >>>>>>>> the failure: >> >>>>>>>> >> >>>>>>>> $ dist/prototype_ia32-linux/rvm -jar ~/tests/dacapo-2006-10- >> >>>>>>>> MR2.jar chart >> >>>>>>>> >> >>>>>>>> ===== DaCapo chart starting ===== >> >>>>>>>> >> >>>>>>>> Fatal error: NullPointerException within uninterruptible region. >> >>>>>>>> Exiting virtual machine due to uninterruptibility violation. >> >>>>>>>> >> >>>>>>>> -- Stack -- >> >>>>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >> >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> >>>>>>>> deliverHardwareException(II)V at line 685 >> >>>>>>>> at <hardware trap> >> >>>>>>>> at Lorg/jikesrvm/runtime/Magic; >> >>>>>>>> >> >>>>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >> >>>>>>>> unboxed/Address; >> >>>>>>>> at line 166 >> >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> >>>>>>>> >> >>>>>>>> >> >>>>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >> >>>>>>>> unboxed/Address; >> >>>>>>>> at line 1003 >> >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >> >>>>>>>> line 61 >> >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >> >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >> >>>>>>>> at Lorg/jikesrvm/classloader/RVMClass; >> >>>>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line >> >>>>>>>> 1119 >> >>>>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >> >>>>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >> >>>>>>>> Address;)I >> >>>>>>>> at line 185 >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> JikesRVM: internal error: recursive use of hardware exception >> >>>>>>>> registers (exiting) >> >>>>>>>> Scheduler.dumpStack(): in a recursive call, 2 deep. >> >>>>>>>> >> >>>>>>>> -- Stack -- >> >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >> >>>>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;Lorg/vmmagic/unboxed/ >> >>>>>>>> Address;)V >> >>>>>>>> at line 732 >> >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >> >>>>>>>> dumpStack(Lorg/vmmagic/unboxed/Address;)V at line 693 >> >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; tracebackWithoutLock()V at >> >>>>>>>> line >> >>>>>>>> 650 >> >>>>>>>> at Lorg/jikesrvm/scheduler/Scheduler; >> >>>>>>>> traceback(Ljava/lang/String;)V at line 628 >> >>>>>>>> at Lorg/jikesrvm/VM; sysFail(Ljava/lang/String;)V at line 2144 >> >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> >>>>>>>> deliverHardwareException(II)V at line 685 >> >>>>>>>> at <hardware trap> >> >>>>>>>> at Lorg/jikesrvm/runtime/Magic; >> >>>>>>>> >> >>>>>>>> getReturnAddress(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >> >>>>>>>> unboxed/Address; >> >>>>>>>> at line 166 >> >>>>>>>> at Lorg/jikesrvm/runtime/RuntimeEntrypoints; >> >>>>>>>> >> >>>>>>>> >> >>>>>>>> unwindNativeStackFrame(Lorg/vmmagic/unboxed/Address;)Lorg/vmmagic/ >> >>>>>>>> unboxed/Address; >> >>>>>>>> at line 1003 >> >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrameInternal(Z)Z at >> >>>>>>>> line 61 >> >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; upOneFrame()V at line 98 >> >>>>>>>> at Lorg/jikesrvm/runtime/StackBrowser; up()V at line 110 >> >>>>>>>> at Lorg/jikesrvm/classloader/RVMClass; >> >>>>>>>> getClassLoaderFromStackFrame(I)Ljava/lang/ClassLoader; at line >> >>>>>>>> 1119 >> >>>>>>>> at Lorg/jikesrvm/jni/JNIFunctions; >> >>>>>>>> FindClass(Lorg/jikesrvm/jni/JNIEnvironment;Lorg/vmmagic/unboxed/ >> >>>>>>>> Address;)I >> >>>>>>>> at line 185 >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> at <native frame> >> >>>>>>>> JikesRVM: internal error: recursive use of hardware exception >> >>>>>>>> registers (exiting) >> >>>>>>>> >> >>>>>>>> I also checked other Dacapo benches and they all were finished >> >>>>>>>> correctly on this prototype configuration. This issue also >> >>>>>>>> reproduces >> >>>>>>>> for me at r15100, r15000, r14950. The cattrack is down, so I >> >>>>>>>> don't >> >>>>>>>> know whether this issue is reproduced on test machines. JIRA >> >>>>>>>> search >> >>>>>>>> for "Dacapo chart" didn't bring me to any of relevant issues. >> >>>>>>>> >> >>>>>>>> Is this new or some known issue? >> >>>>>>>> >> >>>>>>>> Thanks, >> >>>>>>>> Aleksey. >> >>>>>>>> >> >>>>>>>> >> >>>>>>>> ------------------------------------------------------------------------- >> >>>>>>>> This SF.Net email is sponsored by the Moblin Your Move >> >>>>>>>> Developer's >> >>>>>>>> challenge >> >>>>>>>> Build the coolest Linux based applications with Moblin SDK & win >> >>>>>>>> great >> >>>>>>>> prizes >> >>>>>>>> Grand prize is a trip for two to an Open Source event anywhere in >> >>>>>>>> the >> >>>>>>>> world >> >>>>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> >>>>>>>> _______________________________________________ >> >>>>>>>> Jikesrvm-researchers mailing list >> >>>>>>>> Jikesrvm-researchers@... >> >>>>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >>>>>>> >> >>>>>>> >> >>>>>>> >> >>>>>>> ------------------------------------------------------------------------- >> >>>>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >> >>>>>>> challenge >> >>>>>>> Build the coolest Linux based applications with Moblin SDK & win >> >>>>>>> great >> >>>>>>> prizes >> >>>>>>> Grand prize is a trip for two to an Open Source event anywhere in >> >>>>>>> the world >> >>>>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> >>>>>>> _______________________________________________ >> >>>>>>> Jikesrvm-researchers mailing list >> >>>>>>> Jikesrvm-researchers@... >> >>>>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >>>>>>> >> >>>>>>> >> >>>>>> >> >>>>> >> >>>>> >> >>>>> ------------------------------------------------------------------------- >> >>>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >> >>>>> challenge >> >>>>> Build the coolest Linux based applications with Moblin SDK & win >> >>>>> great prizes >> >>>>> Grand prize is a trip for two to an Open Source event anywhere in >> >>>>> the world >> >>>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> >>>>> _______________________________________________ >> >>>>> Jikesrvm-researchers mailing list >> >>>>> Jikesrvm-researchers@... >> >>>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >>>> >> >>>> >> >>>> >> >>>> ------------------------------------------------------------------------- >> >>>> This SF.Net email is sponsored by the Moblin Your Move Developer's >> >>>> challenge >> >>>> Build the coolest Linux based applications with Moblin SDK & win >> >>>> great prizes >> >>>> Grand prize is a trip for two to an Open Source event anywhere in the >> >>>> world >> >>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> >>>> _______________________________________________ >> >>>> Jikesrvm-researchers mailing list >> >>>> Jikesrvm-researchers@... >> >>>> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers >> >>>> >> >>> >> >> >> > >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the >> world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Jikesrvm-researchers mailing list >> Jikesrvm-researchers@... >> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Jikesrvm-researchers mailing list > Jikesrvm-researchers@... > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers > > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jikesrvm-researchers mailing list Jikesrvm-researchers@... https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers |
| Free embeddable forum powered by Nabble | Forum Help |
