John Levon wrote:
> On Sat, Dec 21, 2002 at 01:56:05AM +0100, Alfredo Braunstein wrote:
> I do not want to apply this until you have confirmation from Trolltech
> that this is a bug, a nd the workaround is good.
>
> Send an example app to [EMAIL PROTECTED], they are quite responsive
>
> regards
> john
>
>
So this is what I've got:
>> The following test program compiled with qt 3.0.5 on a Redhat 7.3
>> system results in the following message:
>>
>> Mutex destroy failure: Device or resource busy
>> On exit. Removing the comments avoids the message.
>>
>> Is it a known problem?
>>
>> If it is a qt bug, is this a correct/safe workaround?>
>>
>> If it isn't, what's going on? How can we avoid the message (and
>> possible
>> side-effects, of course). Note that we cannot/want to call a.exec() in
>> our own program.
>>
>> //#define QT_THREAD_SUPPORT
>> #include <qapplication.h>
>>
>> int main(int argc, char **argv)
>> {
>> QApplication a(argc,argv);
>> // a.unlock();
>> }
>>
>>
>> Thank you for your patience,
>>
>> Alfredo Braunstein
>>
>
>Hi Alfredo
>
>This is not a bug and the a.unlock() is a safe and appropriate method to
>avoid the warning.
>
>best regards
>
>Anders Bakken
>--
>Trolltech AS, Waldemar Thranes gate 98, NO-0175 Oslo, Norway
I don't know, however, about the #define QT_THREAD_SUPPORT thing...
are there versions of qt compiled without? Should we detect this?
Bye, Alfredo
