Re: [rvm-research] Help on JikesRVM: internal error: recursive use of hardware exception registers

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 454-5900

Hi Du,

The function called in Read/Write Barrier is "inNursery" from Gen.java

Thanks!
Best
Junjie

________________________________
 From: Du Li <daw...@gm...>
To: Junjie Qian <jun...@ya...>; "General discussion of Jikes RVM design, implementation, issues, and plans" <jik...@li...> 
Sent: Friday, December 9, 2011 10:41 AM
Subject: Re: [rvm-research] Help on JikesRVM: internal error: recursive use of hardware exception registers

which function did you call?

Du

On Fri, Dec 9, 2011 at 10:39 AM, Junjie Qian <jun...@ya...> wrote:

Hi Steve,
>
>
>That is true. 
>I think the problem is that, I call one function from Gen.java in Read/Write Barrier.
>That means, I make one function from Gen.java public, and then call this function in Read/Write Barrier.
>But I have no idea how to fix this, or what should I do to avoid the error.
> 
>Could anyone give me some hints or suggestions on this?
>
>
>Thanks!
>Best
>Junjie
>
>
>The error shows as,
>>>>
>>>>            JikesRVM: internal error: recursive use of hardware
>>>>            exception registers (exiting)
>>>>            -- Stack --
>>>>            at [0x67853e78, 0x64682202]
>>>>     
       Lorg/jikesrvm/runtime/RuntimeEntrypoints;
>>>>            deliverHardwareException(II)V at line 694
>>>>            at [0x67853e94, 0x64682202] <hardware trap>
>>>>            at [0x67859eb4, 0x6468d9a0]
>>>>            Lorg/mmtk/plan/generational/GenCollector; collectionPhase(SZ)V
>>>>            at [0x67859efc, 0x64d78a9e]
>>>>            Lorg/mmtk/plan/generational/immix/GenImmixCollector;
>>>>            collectionPhase(SZ)V at line 168
>>>>            at [0x67859f64, 0x646864ef] Lorg/mmtk/plan/Phase;
>>>>            processPhaseStack(Z)Z at line 429
>>>>            at [0x67859f84, 0x646bbc6c]
 Lorg/mmtk/plan/Phase;
>>>>            beginNewPhaseStack(I)Z at line 351
>>>>            at [0x67859f84, 0x646bbc6c]
>>>>            Lorg/mmtk/plan/StopTheWorldCollector; collect()V at line 39
>>>>            at [0x67859fc8, 0x64eb5ec4]
>>>>            Lorg/jikesrvm/mm/mminterface/CollectorThread; run()V at line 369
>>>>            at [0x6785a000, 0x0804fd2e]
>>>>            Lorg/jikesrvm/scheduler/RVMThread; startoff()V at line 2593
>>>>            Effective sampling rate = 0.01 (10058 (0x000000000000274a) /
>>>>            1047620 (0x00000000000ffc44))
>>>>            Races: 0 distinct, 0 dynamic
>>>>
>>>>   
         JikesRVM: exit 124
>
>
>
>________________________________
> From: Steve Blackburn <Ste...@an...>
>
>To: Junjie Qian <jun...@ya...>; "General discussion of Jikes RVM design, implementation, issues, and plans" <jik...@li...> 
>Sent: Friday, December 9, 2011 12:14 AM
>
>Subject: Re: [rvm-research] Help on JikesRVM: internal error: recursive use of hardware exception registers
>
>
>
>Hi Junjie,
>
>
>If you can reproduce the problem in a version of Jikes RVM that you have not modified, please let us know.
>
>
>Otherwise, it's fair to assume that the problem is due to something you did.   We can't begin to guess what that might be.
>
>
>If you can't reproduce the problem in a regular version of Jikes RVM, the trick is to try to make the smallest change possible to the working version that introduces the problem.
>
>
>--Steve
>
>
>On 09/12/2011, at 5:07 PM, Junjie Qian wrote:
>
>Hi Daniel,
>>
>>
>>Thanks for help!
>>
>>
>>The Jikesrvm has been changed to include some other functions.
>>
>>
>>And
>>"Effective sampling rate = 0.01 (10058 (0x000000000000274a) / 1047620 (0x00000000000ffc44))
>>Races: 0 distinct, 0 dynamic"
>>This information does not show in most outputs.
>> 
>>Thanks!
>>Best
>>Junjie
>>
>>
>>
>>________________________________
>> From: Daniel Frampton <zyr...@zy...>
>>To: Junjie Qian <jun...@ya...>; "General discussion of Jikes RVM design, implementation, issues, and plans" <jik...@li...> 
>>Cc: Robin Garner <rob...@an...> 
>>Sent: Thursday, December 8, 2011 11:59 PM
>>Subject: Re: [rvm-research] Help on JikesRVM: internal error: recursive use of hardware exception registers
>> 
>>
>>Is this an unmodified JikesRVM? I do not recognize:
>> 
>>Effective sampling rate = 0.01 (10058 (0x000000000000274a) / 1047620 (0x00000000000ffc44))
>>Races: 0 distinct, 0 dynamic
>> 
>>From your output.
>> 
>>Cheers,
>>Daniel.
>>
>>On Fri, Dec 9, 2011 at 4:51 PM, Junjie Qian <jun...@ya...> wrote:
>>
>>Hi Robin,
>>>
>>>
>>>Thanks for help!
>>>
>>>
>>>The addresses show with the command ldd is as follows, and I think no one is in the range 0x64000000 to 0xb0000000.
>>>
>>>
>>>        linux-gate.so.1 =>  (0xf7717000)
>>>        librt.so.1 => /lib32/librt.so.1 (0xf76ed000)
>>>        libpthread.so.0 => /lib32/libpthread.so.0 (0xf76d2000)
>>>        libdl.so.2 => /lib32/libdl.so.2 (0xf76cc000)
>>>        librvm.so => not found
>>>        libstdc++.so.6 => /usr/lib32/libstdc++.so.6 (0xf75e1000)
>>>        libm.so.6 => /lib32/libm.so.6 (0xf75b7000)
>>>        libgcc_s.so.1 => /usr/lib32/libgcc_s.so.1 (0xf7599000)
>>>        libc.so.6 => /lib32/libc.so.6 (0xf741e000)
>>>        /lib/ld-linux.so.2 (0xf7718000)
>>>
>>>
>>>There is still same error information with the option "-X:gc:eagerMmapSpaces=true".
>>>Any more suggestions? 
>>>
>>>
>>>Thanks!
>>>Best
>>>Junjie
>>>
>>>
>>>
>>>________________________________
>>> From: Robin Garner <rob...@an...>
>>>
>>>To: Junjie Qian <jun...@ya...>; "General discussion of Jikes RVM design, implementation, issues, and plans" <jik...@li...> 
>>>Sent: Thursday, December 8, 2011 11:07 PM
>>>Subject: Re: [rvm-research] Help on JikesRVM: internal error: recursive use of hardware exception registers
>>>
>>>
>>>Junjie,
>>>
>>>It's possible that Linux has mapped a shared library in the space 
>>>occupied by MMTk's heap.
>>>
>>>If this command
>>>
>>>   ldd dist/production_x86_64-linux/JikesRVM
>>>
>>>shows any addresses in the range 0x64000000 to 0xb0000000, this may be 
>>>your problem.
>>>
>>>You could also try running with the option
>>>
>>>   -X:gc:eagerMmapSpaces=true
>>>
>>>and see if this helps.
>>>
>>>
>>>I expect you're running a production build ?  You could try building a 
>>>development image (bin/buildit localhost development) and see if this 
>>>gievs any more clues.
>>>
>>>cheers
>>>
>>>On 09/12/11 15:52, Junjie Qian wrote:
>>>> Du,
>>>>
>>>> I tried jython and antlr in Dacapo, and both two have same error
>>>> information.
>>>> I run xalan yesterday, and it also has this error.
>>>>
>>>> Thanks!
>>>> Best
>>>> Junjie
>>>>
>>>>
 ------------------------------------------------------------------------
>>>> *From:* Du Li <daw...@gm...>
>>>> *To:* Junjie Qian <jun...@ya...>; "General discussion of Jikes
>>>> RVM design, implementation, issues, and plans"
>>>> <jik...@li...>
>>>> *Sent:* Thursday, December 8, 2011 10:50 PM
>>>> *Subject:* Re: [rvm-research] Help on JikesRVM: internal error:
>>>> recursive use of hardware exception registers
>>>>
>>>> only for this benchmark? how about other benchmark in Dacapo?
>>>>
>>>> Du
>>>>
>>>> On Thu, Dec 8, 2011 at 10:45 PM, Junjie Qian <jun...@ya...
>>>> <mailto:jun...@ya...>> wrote:
>>>>
>>>>     Du,
>>>>
>>>>     I changed the heap size from 90M to 800M (100M each step), and the
>>>>     result shows the same except the execution time difference.
>>>>
>>>>     Thanks!
>>>>     Best
>>>>     Junjie
>>>>
>>>>     ------------------------------------------------------------------------
>>>>     *From:* Du Li <daw...@gm... <mailto:daw...@gm...>>
>>>>     *To:* Junjie Qian <jun...@ya...
>>>>     <mailto:jun...@ya...>>
>>>>     *Cc:* "General discussion of Jikes RVM design, implementation,
>>>>     issues, and plans" <jik...@li...
>>>>     <mailto:jik...@li...>>
>>>>     *Sent:* Thursday, December 8, 2011 10:43 PM
>>>>
>>>>     *Subject:* Re: [rvm-research] Help on JikesRVM: internal error:
>>>>     recursive use of hardware
 exception registers
>>>>
>>>>     what is current heap size now?
>>>>
>>>>     Du
>>>>
>>>>     On Thu, Dec 8, 2011 at 10:41 PM, Junjie Qian <jun...@ya...
>>>>     <mailto:jun...@ya...>> wrote:
>>>>
>>>>         Hi Du,
>>>>
>>>>         The benchmark I am using is Dacapo-2006 single thread benchmarks.
>>>>         This error doesnot happen each time, but most times it happens
>>>>         and the error information are same.
>>>>
>>>>         At first, when I try to increase the heap size, there will be no
>>>>         error, but now it seems that it has
 little to do with the heap
>>>>         size and threads.
>>>>         Any idea?
>>>>
>>>>         Thanks!
>>>>         Best
>>>>         Junjie
>>>>
>>>>         ------------------------------------------------------------------------
>>>>         *From:* Du Li <daw...@gm... <mailto:daw...@gm...>>
>>>>         *To:* Junjie Qian <jun...@ya...
>>>>         <mailto:jun...@ya...>>; "General discussion of Jikes
>>>>         RVM design, implementation, issues, and plans"
>>>>         <jik...@li...
>>>>         <mailto:jik...@li...>>
>>>>         *Sent:* Thursday, December 8, 2011 10:32 PM
>>>>         *Subject:* Re: [rvm-research] Help on JikesRVM: internal error:
>>>>         recursive use of hardware exception registers
>>>>
>>>>         Junjie,
>>>>
>>>>         Which benchmark are you using?
>>>>        
 Does this error happen in each run?
>>>>
>>>>         Thanks.
>>>>
>>>>         Du
>>>>
>>>>         On Thu, Dec 8, 2011 at 10:25 PM, Junjie Qian
>>>>         <jun...@ya... <mailto:jun...@ya...>> wrote:
>>>>
>>>>             Hi guys,
>>>>
>>>>             I am now using Jikesrvm 3.1.0 to run one benchmark, and
>>>>             encounter the following error.
>>>>             I searched through the google, and got some idea how this
>>>>             error comes up, but do not know how to fix
 it.
>>>>
>>>>             Could anyone give me some hint on this?
>>>>
>>>>             Thanks!
>>>>             Best
>>>>             Junjie
>>>>
>>>>             The error shows as,
>>>>
>>>>             JikesRVM: internal error: recursive use of hardware
>>>>             exception registers (exiting)
>>>>             -- Stack --
>>>>             at [0x67853e78, 0x64682202]
>>>>             Lorg/jikesrvm/runtime/RuntimeEntrypoints;
>>>>             deliverHardwareException(II)V at line 694
>>>>             at [0x67853e94, 0x64682202] <hardware
 trap>
>>>>             at [0x67859eb4, 0x6468d9a0]
>>>>             Lorg/mmtk/plan/generational/GenCollector; collectionPhase(SZ)V
>>>>             at [0x67859efc, 0x64d78a9e]
>>>>             Lorg/mmtk/plan/generational/immix/GenImmixCollector;
>>>>             collectionPhase(SZ)V at line 168
>>>>             at [0x67859f64, 0x646864ef] Lorg/mmtk/plan/Phase;
>>>>             processPhaseStack(Z)Z at line 429
>>>>             at [0x67859f84, 0x646bbc6c] Lorg/mmtk/plan/Phase;
>>>>             beginNewPhaseStack(I)Z at line 351
>>>>             at [0x67859f84, 0x646bbc6c]
>>>>            
 Lorg/mmtk/plan/StopTheWorldCollector; collect()V at line 39
>>>>             at [0x67859fc8, 0x64eb5ec4]
>>>>             Lorg/jikesrvm/mm/mminterface/CollectorThread; run()V at line 369
>>>>             at [0x6785a000, 0x0804fd2e]
>>>>             Lorg/jikesrvm/scheduler/RVMThread; startoff()V at line 2593
>>>>             Effective sampling rate = 0.01 (10058 (0x000000000000274a) /
>>>>             1047620 (0x00000000000ffc44))
>>>>             Races: 0 distinct, 0 dynamic
>>>>
>>>>             JikesRVM: exit 124
>>>>
>>>>             ------------------------------------------------------------------------------
>>>>         
    Cloud Services Checklist: Pricing and Packaging Optimization
>>>>             This white paper is intended to serve as a reference,
>>>>             checklist and point of
>>>>             discussion for anyone considering optimizing the pricing and
>>>>             packaging model
>>>>             of a cloud services business. Read Now!
>>>>            http://www.accelacomm.com/jaw/sfnl/114/51491232/
>>>>             _______________________________________________
>>>>             Jikesrvm-researchers mailing list
>>>>            Jik...@li...
>>>>             <mailto:Jik...@li...>
>>>>            https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> This body part will be downloaded on demand.
>>>>
>>>>
>>>>
>>>> This body part will be downloaded on demand.
>>>
>>>
>>>
>>>
>>>------------------------------------------------------------------------------
>>>Cloud Services Checklist: Pricing and Packaging Optimization
>>>This white paper is intended to serve as a reference, checklist and point of
>>>discussion for anyone considering optimizing the pricing and packaging model
>>>of a cloud services business. Read Now!
>>>http://www.accelacomm.com/jaw/sfnl/114/51491232/
>>>_______________________________________________
>>>Jikesrvm-researchers mailing list
>>>Jik...@li...
>>>https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers
>>>
>>>
>>
>>
>>------------------------------------------------------------------------------
>>Cloud Services Checklist: Pricing and Packaging Optimization
>>This white paper is intended to serve as a reference, checklist and point of 
>>discussion for anyone considering optimizing the pricing and packaging model 
>>of a cloud services business. Read Now!
>>http://www.accelacomm.com/jaw/sfnl/114/51491232/_______________________________________________
>>Jikesrvm-researchers mailing list
>>Jik...@li...
>>https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers
>>
>
>
>
>------------------------------------------------------------------------------
>Cloud Services Checklist: Pricing and Packaging Optimization
>This white paper is intended to serve as a reference, checklist and point of
>discussion for anyone considering optimizing the pricing and packaging model
>of a cloud services business. Read Now!
>http://www.accelacomm.com/jaw/sfnl/114/51491232/
>_______________________________________________
>Jikesrvm-researchers mailing list
>Jik...@li...
>https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers
>
>