Hi

On 5 July 2012 08:39, Jan Groenewald <[email protected]> wrote:

> Hi
>
> On 5 July 2012 08:00, Dima Pasechnik <[email protected]> wrote:
>
>>
>>
>> On Thursday, 5 July 2012 00:11:11 UTC+8, Jan Groenewald wrote:
>>>
>>> Hi
>>>
>>> On 4 July 2012 17:46, Dima Pasechnik <[email protected]> wrote:
>>>
>>>> indeed, it seems to be not reproducible on "standard" Sage installs, so
>>>> far (I also tried Linux x86_64 with Sage 5.1.something)
>>>>
>>>>
>>>> On Wednesday, 4 July 2012 18:38:05 UTC+8, Volker Braun wrote:
>>>>>
>>>>> This looks suspiciously like a bug in ATLAS-devel (3.9.x) that I
>>>>> recently helped to track down. Is the Ubuntu PPA compiling Sage's own 
>>>>> ATLAS
>>>>> library or is it using a system library? Note that Sage ships ATLAS-stable
>>>>> (3.8.x) for a good reason.
>>>>
>>>>
>>> Uhm, isn't the stable release compiled upstream at sagemath.org more
>>> "standard" than
>>> a 5.1 beta compiled on a personal computer??
>>>
>>
>> by "standard" I meant a non-PPA install.
>>
>
> I donwloaded the upstream binary and tried that (non-PPA) instead.
> The stable 5.0.1 on Ubuntu 12.04. It is still reproducible.
>
> With the pyc files:
>
>  Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 0x7fffb2ee0700 (LWP 31255)]
> 0x00007fffeb7a9f0d in ATL_dJIK40x40x40TN40x40x0_a1_b1 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> (gdb) bt
> #0  0x00007fffeb7a9f0d in ATL_dJIK40x40x40TN40x40x0_a1_b1 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #1  0x00007fffeb819f8f in ATL_dmmIJK2 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #2  0x00007fffeb81a781 in ATL_dmmIJK ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #3  0x00007fffeb810be5 in ATL_dgemm ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #4  0x00007fffebc3e16d in ATL_dptgemm0 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
>
> #5  0x00007ffff77ece9a in start_thread ()
>    from /lib/x86_64-linux-gnu/libpthread.so.0
> #6  0x00007ffff6e194bd in clone () from /lib/x86_64-linux-gnu/libc.so.6
> #7  0x0000000000000000 in ?? ()
> (gdb)
>
> Without the pyc files (removed them first)
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 0x7fffb2ee0700 (LWP 32264)]
> 0x00007fffeb7a9f0d in ATL_dJIK40x40x40TN40x40x0_a1_b1 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> (gdb) bt
> #0  0x00007fffeb7a9f0d in ATL_dJIK40x40x40TN40x40x0_a1_b1 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #1  0x00007fffeb819f8f in ATL_dmmIJK2 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #2  0x00007fffeb81a781 in ATL_dmmIJK ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #3  0x00007fffeb810be5 in ATL_dgemm ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
> #4  0x00007fffebc3e16d in ATL_dptgemm0 ()
>    from
> /srv/local/Disk_Space/jan/sage-5.0.1-linux-64bit-ubuntu_10.04.3_lts-x86_64-Linux/local/lib/libatlas.so
>
> #5  0x00007ffff77ece9a in start_thread ()
>    from /lib/x86_64-linux-gnu/libpthread.so.0
> #6  0x00007ffff6e194bd in clone () from /lib/x86_64-linux-gnu/libc.so.6
> #7  0x0000000000000000 in ?? ()
> (gdb)
>
> I will try a from-source compile of the stable release next.
>

This seems to be an Ubuntu 12.04 problem. On Ubuntu 11.04
with the same upstream binary:

    5 0.0020140316698371023 0.001542013044808182
   6 0.0019834456743228197 0.0015133962643927208
   7 0.0019536152967604622 0.0014857142847366934
   8 0.0019245234850758745 0.0014589328600844204
   9 0.0018961547284709374 0.0014330137481604196
  10 0.0018684955292540726 0.0014079150575883282
  11 0.0018415347444111442 0.0013835917295943654
  12 0.0018152637651007548 0.0013599961433163651
  13 0.0017896765121765098 0.0013370788289153368
  14 0.001764769237463005 0.0013147892681664715
  15 0.0017405401331557621 0.0012930767585252209
  16 0.0017169887646163316 0.0012718913139809771
  17 0.0016941153542435229 0.0012511845744305283
  18 0.0016719199551473378 0.0012309106948410859
  19 0.0016504015623261317 0.0012110271860831642
  20 0.0016295572152261967 0.0011914956808990625
  21 0.0016093811484123527 0.0011722826008898637
  22 0.0015898640462462818 0.001153359703499124
  23 0.0015709924528396796 0.0011347044915844869
  24 0.0015527483803025274 0.001116300472154339
...


Regards,
Jan


-- 
  .~.
  /V\     Jan Groenewald
 /( )\    www.aims.ac.za
 ^^-^^

-- 
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

Reply via email to