Michael Goffioul wrote
> On Fri, Feb 1, 2013 at 12:11 PM, John W. Eaton <
> jwe@
> > wrote:
>
>> Oops, I forgot to CC the list.
>>
>> jwe
>>
>>
>> -------- Original Message --------
>> Subject: Re: Segfault with Java and "make check"
>> Date: Fri, 01 Feb 2013 11:07:12 -0600
>> From: John W. Eaton <
> jwe@
> >
>> To: Rik <
> rik@
> >
>> CC:
> jwe@
>>
>> On 02/01/2013 10:40 AM, Rik wrote:
>>
>> This seems like a reasonable supposition. The quadcc algorithm uses
>>> adaptive rules which involve recursively subdividing the integration
>>> region. When the function is smooth the process doesn't go very far.
>>> Around singularities or discontinuities the algorithm has to shrink the
>>> integration step size to small values to ensure it maintains the overall
>>> accuracy desired. This would mean lots of recursion.
>>>
>>
>> The JVM is not supposed to be loaded unless it is actually needed. Is
>> there some test that runs prior to the one that fails that causes the
>> JVM to be loaded?
>
>
> Yes, I guess there is. There are java-related tests in
> libinterp/octave-value/ov-java.cc, and these are run before reaching the
> tests in m-files.
(Indeed. I've put those tests in, on Rik's request, and I have a few more in
various states of development pending)
Isn't there a way to unload the JVM and free its resources if it isn't
needed anymore, either manually, or automatically after a period of non-use?
This is explained here [1]. What I don't know is whether you'll be able to reload the JVM cleanly later on. I never tried that.
Michael.