On 04/30/2012 10:50 AM, Serghei Amelian wrote:
> On Monday 30 April 2012 18:39:59 David C. Rankin wrote:
>> Tim, Darrell,
>>
>>     As /dev/ammo42 suggested, I captured a strace on the i686 crash. It is
>> attached to bug 979. (current tdebase and tqt3, tdelibs from 4/16 -- same
>> crash)
>>
>> http://bugs.trinitydesktop.org/attachment.cgi?id=568
>>
>>     Also, this bug occurs 100% of the time when pasting 'small' amounts of
>> text (1-2 words at a time) until the line wraps. I can paste 10 lines of
>> text into kwrite and keep editing - so it looks like something that effects
>> the single line length computation.
>>
>>     I'll keep incrementally downgrading and report when I narrow the time
>> the bug was introduced. If anyone can read straces, then take a look.
>
> Be very carefully with gcc 4.7 patches. For example, if you have something
> like this:
>
> Iterator it = blabla;
> for(Iterator it = blabla; it; ++it)
> 	it.do_something()
>
> Is not enough to rename the for's "it", this will fix building problem, but
> will broke the logic of loop. You should also rename the "it" which is inside
> the loop.
>
>
> This is WRONG:
>
> Iterator it = blabla;
> for(Iterator it1 = blabla; it1; ++it1)
> 	it.do_something()
>
>
>
> This is CORRECT:
>
> Iterator it = blabla;
> for(Iterator it1 = blabla; it1; ++it1)
> 	it1.do_something()
>
>

Serghei,

   Thank you. I was pretty careful there, but I can always screw something up 
:) However, so far I haven't patched anything in tqt3, tqtinterface, kdelibs 
or kdebase (aside from the easyvector kicker patch and the konsole tips patch 
-- that involved no logic change at all) related to iterators. Those were to 
rosegarden and other apps.

   I think we are good there. You got any other guesses about this crash? You 
guys seeing this in openSuSE factory yet?


-- 
David C. Rankin, J.D.,P.E.