[kde-edu]: Step segfault

[Vladimir Kuznetsov]

Hello,

Yes, valgrind may give you a useful information (or may not - it depends).

Unfortunately I am unable to reproduce the problem on my PC. What
versions of KDE and Qt are you using ?

-- 
Best regards,
Vladimir Kuznetsov



On Wed, Dec 16, 2009 at 3:58 PM, Glen Kaukola <gkaukola at cs.ucr.edu> wrote:
>> As Lück already said, the problem can no longer be reproduced in KDE
>> 4.3. Could you try the resent version ?
>
> I'm going with trunk.  I suppose I might give it a go on openSUSE as
> opposed to Fedora just to see where that gets me.
>
>> How have you found that the problem is in Number::floatValue() ? Is it
>> crashed inside that function ?
>> If so, it might be caused also by wrong arguments passed to that
>> function, or by previous memory corruption.
>
> Well, it goes UnitsCalc::parseNumber() -> Number::floatValue() ->
> cln::double_approx().  So I guess the problem is potentially in cln.  But
> like I say, I copied the code over almost verbatim to my test app and the
> test app runs fine, so I'm pretty sure it's not wrong arguments being
> passed.  In fact I'm spitting out debug statements in both the real code
> and the test code and the contents of all the variables seem to match
> exactly.  Even the type of the expr MathStructure.
>
> Would valgrind be the thing to go with to check for previous memory
> corruption?
>
> Thanks,
> Glen Kaukola
>
> _______________________________________________
> kde-edu mailing list
> kde-edu at mail.kde.org
> https://mail.kde.org/mailman/listinfo/kde-edu
>