[PD-dev] Malloc Error occuring with PD 0.41.4

james murray james.n.murray at gmail.com
Fri Aug 28 21:52:36 CEST 2009 

Well,

It steadily creeps up in memory used until I have approx 140 mb of free
memory on the computer.  And although I noticed that it occasionally flashed
all the way down to only 6 - 9 MB of free memory at some points for a
flickering moment, when it finally crashed it was listing free memory at
around 133 mb.  It took almost two hours for this to occur.  I should note
that the Pd-extended 0.41.4 version that is running the camera
input/record/pix_share_write runs steadily at 139 MB both during and after
the crash.  In fact nothing but the version of Pure Data running video
playback crashes....  So, not really sure where things can go from here.
I'm really frustrated that this is happening since PD has always been my go
to performer.

J

On Fri, Aug 28, 2009 at 1:13 PM, Hans-Christoph Steiner <hans at at.or.at>wrote:

>
> Check it right before it crashes.  If its not close to 3gb, then its not
> the 32/64-bit issue that I mentioned.  Since the error mentions std::bad_alloc,
> which points to C++.  I'm guessing it might be a Gem bug then.
>
> .hc
>
> On Aug 28, 2009, at 12:10 PM, james murray wrote:
>
> Hi,
>
>  I'm generally around 1.34 GB real memory for PD-Extended running video
> playback and 139.81 MB real memory for the camera input patch.  (which of
> course means that I have two versions of PD-Extended running, both are
> 0.41.4 PD-Extended).  I should note that I've got pix_share_read and write
> objects moving video from one copy of PD to the other... running it into a
> buffer on the playback patch.
>
> Virtual memory is 1.52 GB for one version of PD, while the other is at 1.19
> GB VM... I'm running one version of PD for the camera and one for the
> playback.  If other info is needed, I'm happy to supply it.
>
> Thanks
> J
>
> On Fri, Aug 28, 2009 at 10:56 AM, Hans-Christoph Steiner <hans at at.or.at>wrote:
>
>>
>> Did you run Activity Monitor to see how much memory 'pd' is using?
>>
>> .hc
>>
>> On Aug 28, 2009, at 10:48 AM, james murray wrote:
>>
>> I let PD-extended 0.41.4 run again the programming with a reduction from 4
>> GB of memory to 2 GM of memory... and it returned the same error on both
>> machines... I'm still getting the malloc error.... I'm not really sure what
>> options there are....  Does anyone have ideas about this being a 64 bit to
>> 32 bit issue?  Is there something else to try?  I'm at wits end with PD
>> constantly crashing and while competent in programming, malloc errors are
>> definitely beyond my scope of expertise or knowledge.  Any help would be
>> huge.
>>
>> Again, I'm running PD Extended 0.41.4 on It happens every time without
>> variation... I've tried it on two machines, one is:
>>
>> MacPro Quad-Core Intel Xeon, 2.93 GHz processor, 4 GB memory, ATI Radeon
>> 4870 (the apple tweaked version, not the ATI version), BlackMagic Intensity
>> Pro video card... and it's mac 10.5.8.
>>
>> The other computer is identical, but is running on 10.5.6.
>>
>> The error I get is:
>>
>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]
>> pd(272,0xa03ce720) malloc: *** mmap(size=4149248) failed (error code=12)
>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]    ***
>> error: can't allocate region
>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]    *** set a
>> breakpoint in malloc_error_break to debug
>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]    terminate
>> called after throwing an instance of 'std::bad_alloc'
>>
>>
>> Thanks in advance
>> J
>>
>>
>> On Thu, Aug 27, 2009 at 5:46 PM, Hans-Christoph Steiner <hans at at.or.at>wrote:
>>
>>>
>>> FYI, you'll want to keep this on the pd-dev list.  There are others there
>>> who know more about it than me.
>>>
>>> .hc
>>>
>>> On Aug 27, 2009, at 4:16 PM, james murray wrote:
>>>
>>> Hi,
>>>
>>> I switched out the RAM and replaced the 4 GB with two 1 GB modules.
>>> Unfortunately the same crash happen with the same malloc error.  So, not
>>> sure what that means entirely, but, unfortunately it appears to still be an
>>> issue.
>>>
>>> J
>>>
>>> On Thu, Aug 27, 2009 at 12:53 PM, james murray <james.n.murray at gmail.com
>>> > wrote:
>>>
>>>> That makes sense... I'll give it a try and sort it out and see what
>>>> happens...  Thanks for the insight and clarity.
>>>>
>>>> J
>>>>
>>>>
>>>> On Thu, Aug 27, 2009 at 12:45 PM, Hans-Christoph Steiner <hans at at.or.at
>>>> > wrote:
>>>>
>>>>>
>>>>> A 32-bit computer can only handle 3 gigs of RAM.  That's the big deal
>>>>> with 64-bit computers, they handle a lot more.  While your computer is
>>>>> 64-bit and your OS is 64-bit, the build of Pd you are running is 32-bit.
>>>>>  That means that your Pd cannot use more than 3 gigs of RAM no matter how
>>>>> much is in the machine.
>>>>>
>>>>> My guess is that your Pd patch wants to use more than 3gb of RAM, then
>>>>> crashes.  Try running Activity Monitor and seeing how much memory 'pd' is
>>>>> using.
>>>>>
>>>>> I could be wrong that this is the cuase of hte problem tho...
>>>>>
>>>>> .hc
>>>>>
>>>>> On Aug 27, 2009, at 1:41 PM, james murray wrote:
>>>>>
>>>>>
>>>>> Okay... I'd run across this before when trying to research this
>>>>> issue... Sorry to be so dull-witted about this, but does this mean a
>>>>> physical reduction of my RAM itself from 4 GB to 3 GB? Or am I missing the
>>>>> point and I need to do something else?
>>>>>
>>>>>
>>>>> Thanks again for looking into this so quickly, I really appreciate it,
>>>>> getting confirmation about all of this is really huge.
>>>>>
>>>>> J
>>>>>
>>>>> On Thu, Aug 27, 2009 at 12:27 PM, Hans-Christoph Steiner <
>>>>> hans at at.or.at> wrote:
>>>>>
>>>>>>
>>>>>> So it is not the GUI, but 'pd' itself having the issue:
>>>>>> pd(272,0xa03ce720)  My guess is that maybe its a 32-bi/64-bit issue.  I
>>>>>> think that a 32-bit app can only address 3 gigs of RAM.  Pd is buildable has
>>>>>> 64-bit, but there isn't a release for Mac OS X built as 64-bit yet.
>>>>>>
>>>>>> .hc
>>>>>>
>>>>>>
>>>>>> On Aug 27, 2009, at 12:53 PM, james murray wrote:
>>>>>>
>>>>>> It happens every time without variation... I've tried it on two
>>>>>> machines, one is:
>>>>>>
>>>>>> MacPro Quad-Core Intel Xeon, 2.93 GHz processor, 4 GB memory, ATI
>>>>>> Radeon 4870 (the apple tweaked version, not the ATI version), BlackMagic
>>>>>> Intensity Pro video card... and it's mac 10.5.8.
>>>>>>
>>>>>> The other computer is identical, but is running on 10.5.6.
>>>>>>
>>>>>> Thanks for getting back to me s quickly... I've been definitely
>>>>>> hitting my head on this for a few days.
>>>>>>
>>>>>> J
>>>>>>
>>>>>> On Thu, Aug 27, 2009 at 11:42 AM, Hans-Christoph Steiner <
>>>>>> hans at at.or.at> wrote:
>>>>>>
>>>>>>>
>>>>>>> Hmm, never seen that one.  Can you reproduce it, does it happen every
>>>>>>> time? The "org.puredata.pd.wish" part makes me think that make its the
>>>>>>> Tcl/Tk GUI side that's crashing.
>>>>>>>
>>>>>>> Which exact version of mac os x are you running?  Which hardware?
>>>>>>>
>>>>>>> .hc
>>>>>>>
>>>>>>> On Aug 27, 2009, at 12:27 PM, james murray wrote:
>>>>>>>
>>>>>>> Hi List,
>>>>>>>
>>>>>>> I'm currently running into issues with getting PD Extended 0.41.4 to
>>>>>>> run on Leopard 10.5. If I let the software run for an extended period of
>>>>>>> time (say, 30 minutes to a couple of hours) PD/Gem crashes without an error
>>>>>>> report.  I'm running 1920 x 1080 video playback when this happens and I've
>>>>>>> tested the programming with both two and three layers of video running.
>>>>>>>
>>>>>>> This is happening on a quad core Intel-Xeon Mac with 4 GB of memory
>>>>>>> (although the same issue is happening if I cannibalize an additional 4 GB),
>>>>>>> an ATI Radeon card and Blackmagic Intensity for capture.  While I can't get
>>>>>>> an error report from PD/Gem, Console is giving me the following:
>>>>>>>
>>>>>>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]
>>>>>>> pd(272,0xa03ce720) malloc: *** mmap(size=4149248) failed (error code=12)
>>>>>>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]    ***
>>>>>>> error: can't allocate region
>>>>>>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]    ***
>>>>>>> set a breakpoint in malloc_error_break to debug
>>>>>>> 8/25/09 1:38:48 PM    [0x0-0x18018].org.puredata.pd.wish[271]
>>>>>>> terminate called after throwing an instance of 'std::bad_alloc'
>>>>>>>
>>>>>>> This is definitely not an area that I'm overly familiar/confident
>>>>>>> about and would really love to have at least some sort of idea of what could
>>>>>>> be causing this to happen.  This error is the only error I get with my
>>>>>>> programming and it's frustrating since I can't downgrade to Leopard or run
>>>>>>> it on another operating system.  Any insight into this issue would be
>>>>>>> greatly appreciated.  I've altered my programming dramatically and stripped
>>>>>>> it down to the bare bones and still, I have this error occuring.
>>>>>>>
>>>>>>> So, I suppose my question is this... is this a PD/Gem issue with
>>>>>>> Leopard?  Or something on my side that I'm just not seeing... Any insight
>>>>>>> into this would be hugely appreciated....
>>>>>>>
>>>>>>> Thank you,
>>>>>>> J
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Pd-dev mailing list
>>>>>>> Pd-dev at iem.at
>>>>>>> http://lists.puredata.info/listinfo/pd-dev
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> ----------------------------------------------------------------------------
>>>>>>>
>>>>>>> News is what people want to keep hidden and everything else is
>>>>>>> publicity.          - Bill Moyers
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> ----------------------------------------------------------------------------
>>>>>>
>>>>>>                   ¡El pueblo unido jamás será vencido!
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ----------------------------------------------------------------------------
>>>>>
>>>>> I have the audacity to believe that peoples everywhere can have three
>>>>> meals a day for their bodies, education and culture for their minds, and
>>>>> dignity, equality and freedom for their spirits.      - Martin Luther
>>>>> King, Jr.
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> If nature has made any one thing less susceptible than all others of
>>> exclusive property, it is the action of the thinking power called an idea,
>>> which an individual may exclusively possess as long as he keeps it to
>>> himself; but the moment it is divulged, it forces itself into the possession
>>> of everyone, and the receiver cannot dispossess himself of it.            -
>>> Thomas Jefferson
>>>
>>>
>>>
>> _______________________________________________
>> Pd-dev mailing list
>> Pd-dev at iem.at
>> http://lists.puredata.info/listinfo/pd-dev
>>
>>
>>
>>
>>
>> ----------------------------------------------------------------------------
>>
>> Terrorism is not an enemy.  It cannot be defeated.  It's a tactic.  It's
>> about as sensible to say we declare war on night attacks and expect we're
>> going to win that war.  We're not going to win the war on terrorism.
>>   - retired U.S. Army general, William Odom
>>
>>
>>
>
>
>
>
> ----------------------------------------------------------------------------
>
> Looking at things from a more basic level, you can come up with a more
> direct solution... It may sound small in theory, but it in practice, it can
> change entire economies.     - Amy Smith
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.puredata.info/pipermail/pd-dev/attachments/20090828/ca3f381a/attachment.htm>

More information about the Pd-dev mailing list