Hi,
Thanks Paul & Surya for the tip. I was able to run blackscholes without any
issues using the commandline mentioned in create_checkpoint.py script,
without using the {start|stop}_sim. I am hoping I will be able to run other
benchmarks as well.
Regards,
Ankita
On Thu, Sep 13, 2012 at 12:03 PM, snataraj <[email protected]> wrote:
> On 13/09/12 18:08, avadh patel wrote:
>
> This issue can be due to corrupted disk image. Do you use -snapshot option
> while running simulations? Also can you check if you can run those
> checkpoints in emulated mode also?
>
> - Avadh
>
> On Thu, Sep 13, 2012 at 8:22 AM, Alireza Haghdoost
> <[email protected]>wrote:
>
>> That is interesting because I do have the same issue. I mean Assertion
>> failure on page_fault_addr != 0 in my own workload. I thought it might be
>> my workload bug but it seems to me that it might not ... because PARSEC
>> also has this issue as well.
>> Is there any reason in the simulator to receive a page fault on address 0
>> ?
>>
>> Thanks
>> Alireza
>>
>> On Thu, Sep 13, 2012 at 10:15 AM, Paul Rosenfeld <[email protected]>wrote:
>>
>>> You shouldn't need the start and stop sim calls since the roi variant of
>>> parsec already has them built into the binaries. Try to run without them
>>> and see if it still fails.
>>>
>>>
>>> On Thu, Sep 13, 2012 at 10:10 AM, Ankita (Garg) Goel <
>>> [email protected]> wrote:
>>>
>>>> Yes, qemu terminates.
>>>>
>>>> On Thu, Sep 13, 2012 at 8:39 AM, Alireza Haghdoost <[email protected]
>>>> > wrote:
>>>>
>>>>> What happen after receiving this error ? Does it result terminating
>>>>> Qemu ?
>>>>>
>>>>> On Wed, Sep 12, 2012 at 8:37 PM, Ankita (Garg) Goel <
>>>>> [email protected]> wrote:
>>>>>
>>>>>> I am actually not able to run any of the PARSEC benchmarks due to
>>>>>> some errors. When running canneal, the error I hit is:
>>>>>>
>>>>>> ptlsim/build/core/ooo-core/ooo.cpp:1338: bool
>>>>>> ooo::ThreadContext::handle_exception(): Assertion `ctx.page_fault_addr !=
>>>>>> 0' failed.
>>>>>>
>>>>>> The commandline I have used is:
>>>>>>
>>>>>> # <start_sim; parsecmgmt -a run -c gcc-hooks -x roi -n 1 -i
>>>>>> simmedium -p canneal; <stop_sim>
>>>>>>
>>>>>> Any thoughts on this ? Am I missing something ? It is crucial for
>>>>>> me to run some of these multi-threaded benchmarks..
>>>>>>
>>>>>> On Wed, Sep 12, 2012 at 3:53 PM, Ankita (Garg) Goel <
>>>>>> [email protected]> wrote:
>>>>>>
>>>>>>> Hi folks,
>>>>>>>
>>>>>>> Just wondering if anyone has come across the following error when
>>>>>>> running PARSEC blackscholes benchmark ?
>>>>>>>
>>>>>>> qemu-system-x86_64: ptlsim/build/core/ooo-core/ooo-pipe.cpp:2029:
>>>>>>> int ooo::ReorderBufferEntry::commit(): Assertion `physreg->data' failed.
>>>>>>>
>>>>>>> --
>>>>>>> Regards,
>>>>>>> Ankita
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Regards,
>>>>>> Ankita
>>>>>> Graduate Student
>>>>>> Department of Computer Science
>>>>>> University of Texas at Austin
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> http://www.marss86.org
>>>>>> Marss86-Devel mailing list
>>>>>> [email protected]
>>>>>> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Regards,
>>>> Ankita
>>>> Graduate Student
>>>> Department of Computer Science
>>>> University of Texas at Austin
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> http://www.marss86.org
>>>> Marss86-Devel mailing list
>>>> [email protected]
>>>> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>>>>
>>>>
>>>
>>
>> _______________________________________________
>> http://www.marss86.org
>> Marss86-Devel mailing list
>> [email protected]
>> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>>
>>
>
>
> _______________________________________________http://www.marss86.org
> Marss86-Devel mailing
> [email protected]https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>
> did you try running the steps as shown in checkpoint.py like exporting
> the library and running the full syntax instead of parsecmgmt? parsecmgmt
> dint work fr me too but I use the full command line and all the benchmakrs
> runs without anyissue.
>
> --
> Surya Narayanan,
> E301 - 12,
> Inria Rennes - Bretagne Atlantique Research Centre
> Campus universitaire de Beaulieu
> 35042 Rennes Cedex
>
>
> _______________________________________________
> http://www.marss86.org
> Marss86-Devel mailing list
> [email protected]
> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>
>
--
Regards,
Ankita
Graduate Student
Department of Computer Science
University of Texas at Austin
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
