Re: 2.9.14+ hold issue

[David Bateman]

Fredrik Lingvall wrote:
> David Bateman wrote:
>>
>> I think you'll need to supply a simpler example demonstrating this to be
>> able to address it.
>>
>>   
> I have attached a simpler script (testit.m) that triggers the problem
> on my
> 4GB (eg., 3.3 GB) core2 (Dell M90). You need about 3GB ram to run
> this.  The problem
> is however not triggered on an 2 cpu dual core Opteron with 16GB ram.
>
> If it works correctly then octave should complain of the variable Ro_orig
> being undefined and not print:
>
I can't run the script, but not for the reason you can't.. Basically the
code runs for 2 minutes of cpu time on my machine, but uses so much
memory that it swaps horribly. The problem is that the the Out of Memory
manager of the linux kernel then doesn't allow the gnuplot process to
start and I can't test the example.. Could you simplify the example
further to use less memory?

D.


-- 
David Bateman                                address@hidden
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary