[yt-users] problem with slices in non-unitary volumes

Nathan Goldbaum nathan12343 at gmail.com
Wed Dec 23 15:17:56 PST 2015 

I get the same error over here. I think it's unrelated to the other error.
The relevant portion of the C stack trace is here:

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   libsystem_kernel.dylib         0x00007fff8b615286 __pthread_kill + 10
1   libsystem_c.dylib             0x00007fff8bc2a9ab abort + 129
2   libsystem_malloc.dylib         0x00007fff915d5fe2 szone_error + 625
3   libsystem_malloc.dylib         0x00007fff915cf374
small_free_list_remove_ptr + 154
4   libsystem_malloc.dylib         0x00007fff915cac3f
szone_free_definite_size + 1856
5   multiarray.so                 0x00000001044c2d16 npy_free_cache + 70
6   multiarray.so                 0x00000001044c6ed6 array_dealloc + 182
7   selection_routines.so         0x00000001063ffe34
__pyx_pw_2yt_8geometry_18selection_routines_13SliceSelector_3fill_mask +
8804 (selection_routines.c:19749)

Line 19749 of selection_routines.c corresponds to line 1158 in
selection_routines.pyx, which is the place where the fill_mask member
variable of the SliceSelector extension type gets defined.

Unfortunately, I have no idea how to debug this from here...

-Nathan

On Wed, Dec 23, 2015 at 4:57 PM, Brian O'Shea <bwoshea at gmail.com> wrote:

> Actually, interesting followup here - I have to take it back.  When I do a
> projection plot:
>
>     yt.ProjectionPlot(ds,'x',("gas",
> "temperature"),center=[0.0,0.0,0.0],width=(0.2,'Mpc')).save()
>
> the behavior is perfectly fine and I get projections that look sensible.
>  *however*, when I do a slice plot:
>
>    yt.SlicePlot(ds,'x',("gas",
> "temperature"),center=[0.0,0.0,0.0],width=(0.2,'Mpc')).save()
>
> I get a malloc error:
>
> yt : [INFO     ] 2015-12-23 17:50:59,082 Gathering a field list (this may
> take a moment.)
> yt : [INFO     ] 2015-12-23 17:50:59,155 Loading field plugins.
> yt : [INFO     ] 2015-12-23 17:50:59,155 Loaded angular_momentum (8 new
> fields)
> yt : [INFO     ] 2015-12-23 17:50:59,156 Loaded astro (16 new fields)
> yt : [INFO     ] 2015-12-23 17:50:59,156 Loaded cosmology (23 new fields)
> yt : [INFO     ] 2015-12-23 17:50:59,156 Loaded fluid (64 new fields)
> yt : [INFO     ] 2015-12-23 17:50:59,157 Loaded fluid_vector (96 new
> fields)
> yt : [INFO     ] 2015-12-23 17:50:59,157 Loaded geometric (112 new fields)
> yt : [INFO     ] 2015-12-23 17:50:59,157 Loaded local (112 new fields)
> yt : [INFO     ] 2015-12-23 17:50:59,157 Loaded magnetic_field (120 new
> fields)
> yt : [INFO     ] 2015-12-23 17:50:59,158 Loaded my_plugins (120 new fields)
> yt : [INFO     ] 2015-12-23 17:50:59,158 Loaded species (122 new fields)
> python2.7(12254,0x7fff72d19310) malloc: *** error for object 0x103e39008:
> incorrect checksum for freed object - object was probably modified after
> being freed.
> *** set a breakpoint in malloc_error_break to debug
> Abort trap: 6
>
> This happens on my Mac (OS X 10.9.5, LLVM v6.0) and also on my Linux
> machine (Ubuntu 13.04, gcc 4.6.3), although on the linux machine I just get
> a seg fault instead of the malloc stuff.  Using pdb (python -m pdb
> crashing_script.py) gives me the same error with no additional information.
>
>
> On Wed, Dec 23, 2015 at 5:40 PM, Brian O'Shea <bwoshea at gmail.com> wrote:
>
>> Bugfix confirmed; this works for me as well.  Thank you very much!
>>
>> On Wed, Dec 23, 2015 at 5:18 PM, Nathan Goldbaum <nathan12343 at gmail.com>
>> wrote:
>>
>>> I've opened a pull request that allows me to load your dataset:
>>>
>>>
>>> https://bitbucket.org/yt_analysis/yt/pull-requests/1915/bugfix-avoid-creating-particle-unions-for/diff
>>>
>>> On Wed, Dec 23, 2015 at 4:08 PM, Brian O'Shea <bwoshea at gmail.com> wrote:
>>>
>>>> Oh, interesting.  Is there a way to circumvent auto-detection of fields?
>>>>
>>>> On Wed, Dec 23, 2015 at 4:35 PM, Nathan Goldbaum <nathan12343 at gmail.com
>>>> > wrote:
>>>>
>>>>> It looks like field detection is breaking for this Enzo-3.0 active
>>>>> particle dataset. StoredEnergy is available on-disk as ('AGNParticle,
>>>>> 'StoredEnergy'), but for some reason the field detection system isn't able
>>>>> to guess that.
>>>>>
>>>>> On Wed, Dec 23, 2015 at 3:19 PM, Nathan Goldbaum <
>>>>> nathan12343 at gmail.com> wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>> On Wed, Dec 23, 2015 at 11:20 AM, Brian O'Shea <bwoshea at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Dear yt-users,
>>>>>>>
>>>>>>> I'm having a problem with making slices in non-unitary volumes.  I
>>>>>>> have a 3D enzo simulation that has domain bounds of (-1.6...1.6) in every
>>>>>>> dimension (units of megaparsecs), and after I updated to the tip of yt-3
>>>>>>> (changeset d9502d848029) from a fairly old version of the code I suddenly
>>>>>>> have problems with making simple projections.  If I run this set of
>>>>>>> commands:
>>>>>>>
>>>>>>>
>>>>>> Any idea which version you updated from? I tried on yt-3.1 and get
>>>>>> the same error. This is just so we can bisect and find the change that
>>>>>> introduced the regression.
>>>>>>
>>>>>>
>>>>>>> import yt
>>>>>>> ds = yt.load("DD0400/DD0400")
>>>>>>> proj =
>>>>>>> yt.SlicePlot(ds,'x','temperature',center=[0.0,0.0,0.0],width=(0.2,'Mpc'))
>>>>>>>
>>>>>>> I get this error:
>>>>>>>
>>>>>>> http://paste.yt-project.org/show/6115/
>>>>>>>
>>>>>>> This seems to work fine on an Enzo cosmological dataset (domain
>>>>>>> bounds 0-1 in each dimension), and previously worked fine in yt-3 (using a
>>>>>>> version of the code from a couple of months ago).
>>>>>>>
>>>>>>> is this a bug, or am I doing something dumb?
>>>>>>>
>>>>>>> Thanks!
>>>>>>>
>>>>>>> --Brian
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> yt-users mailing list
>>>>>>> yt-users at lists.spacepope.org
>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> yt-users mailing list
>>>>> yt-users at lists.spacepope.org
>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> yt-users mailing list
>>>> yt-users at lists.spacepope.org
>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>
>>>>
>>>
>>> _______________________________________________
>>> yt-users mailing list
>>> yt-users at lists.spacepope.org
>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>
>>>
>>
>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.spacepope.org/pipermail/yt-users-spacepope.org/attachments/20151223/0739ada6/attachment.htm>
-------------- next part --------------
_______________________________________________
yt-users mailing list
yt-users at lists.spacepope.org
http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

More information about the yt-users mailing list