<p dir="ltr"><br>
On Jul 4, 2014 8:33 AM, smc smc <smc101979@gmail.com> wrote:<br>
><br>
> Hi, thanks all for your suggestions.<br>
><br>
> @Sebastian: the bsp is MPPA.<br>
> @Gedare: I cannot send the full code for debugging, because part of it uses an proprietary API :/ About your comment, I commit a mistake transcribing the code, it is not an assignment operator, it must be '=='.<br>
<br>
You should be able to produce a generic version that bus smaller and reproduces the problem.</p>
<p dir="ltr">You did not answer if you have default attributes on the pthreads and other objects. They are undefined by POSIX and differ between RTEMS and Linux. </p>
<p dir="ltr">Also a thread spinning in a loop at the highest priority in RTEMS will do that forever. My impression is that is what your init thread is doing.</p>
<p dir="ltr">> Anyway, I am trying to use condition variables, despite of loosing some of performance.<br>
><br>
> Regards.<br>
><br>
><br>
><br>
> 2014-07-03 2:37 GMT+01:00 Gedare Bloom <gedare@rtems.org>:<br>
>><br>
>> I'll make one note below, but you should submit the entire<br>
>> self-contained example that shows the incorrect behavior if you want<br>
>> debugging help.<br>
>><br>
>> On Wed, Jul 2, 2014 at 6:23 AM, smc smc <smc101979@gmail.com> wrote:<br>
>> > Hi, I am new in RTEMS. I can't understand why the following threaded<br>
>> > application is not running on RTEMS as expected. Basically, it creates two<br>
>> > pthreads. One fills a buffer and the other one get an slice and submit it to<br>
>> > a cluster for processing (typical producer-consumer problem), something<br>
>> > like:<br>
>> ><br>
>> > void thread1()<br>
>> > {<br>
>> > while(1)<br>
>> > {<br>
>> > for (i=0; i < size;i++)<br>
>> > {<br>
>> > if (buffer[i].st == -1)<br>
>> > {<br>
>> > pthread_mutex_lock(&mutex);<br>
>> > genData(buffer[i]);<br>
>> > buffer[i].st=0; // ready to be consumed<br>
>> > pthread_mutex_unlock(&mutex);<br>
>> > }<br>
>> > }<br>
>> > }<br>
>> > }<br>
>> ><br>
>> > void thread2()<br>
>> > {<br>
>> > while(1)<br>
>> > {<br>
>> > // n stores the slice start<br>
>> > ready = false;<br>
>> > while (!ready) // it gets a free slice<br>
>> > {<br>
>> > ready = true;<br>
>> > for(i=n;i<n+slice;i++)<br>
>> > if (buffer[i].st = -1)<br>
>> If you have this statement with an assignment operator, the<br>
>> while(!ready) loop should never terminate if slice > 0.<br>
>><br>
>> > ready = false;<br>
>> > }<br>
>> > // process the slice on cluster<br>
>> > pthread_mutex_lock(&mutex);<br>
>> > for (i=n;i<n+slice;i++)<br>
>> > buffer[i].st = -1; // now new data can be inserted on it<br>
>> > pthread_mutex_unlock(&mutex);<br>
>> > }<br>
>> > }<br>
>> ><br>
>> > It runs fine on a Linux/GNU system, but not in RTEMS. Under some<br>
>> > circumstances the program is blocked. It is like if thread2 was running<br>
>> > forever on the second while, and thread1 didn't fill the buffer. Do you have<br>
>> > any idea of what is happening? I would expect that each thread would run on<br>
>> > parallel on each core.<br>
>> ><br>
>> > Thanks.<br>
>> ><br>
>> > _______________________________________________<br>
>> > users mailing list<br>
>> > users@rtems.org<br>
>> > http://lists.rtems.org/mailman/listinfo/users<br>
><br>
><br>
</p>