<?xml version="1.0"?><!-- generator="bbPress" -->

<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
>

<channel>
<title>Trusster Forums &#187; Topic: Active HDL (Windows)</title>
<link>http://forums.trusster.com/</link>
<description>Trusster Forums &#187; Topic: Active HDL (Windows)</description>
<language>en</language>
<pubDate>Tue, 07 Feb 2012 07:26:08 +0000</pubDate>

<item>
<title>Mike Mintz on "Active HDL (Windows)"</title>
<link>http://forums.trusster.com/topic/active-hdl-windows#post-31</link>
<pubDate>Tue, 04 May 2010 07:33:20 +0000</pubDate>
<dc:creator>Mike Mintz</dc:creator>
<guid isPermaLink="false">31@http://forums.trusster.com/</guid>
<description>&#60;p&#62;Hi Rolf,&#60;/p&#62;
&#60;p&#62;Thanks for trying Teal/Truss and working on these run time issues.  Yeah the synch and mutex code is just ugly.&#60;/p&#62;
&#60;p&#62;I have good contacts at Aldec and will get them to try teal on windows. I know they use teal on Linux in their regression suite.&#60;/p&#62;
&#60;p&#62;If you can get me a stack dump of the crash I will take a look.  &#60;/p&#62;
&#60;p&#62;As for truss, can you get me the error from the compiler. I agree that I made a bad assumption that pthread_t was an integral type.  You can see in the run_loop class (just below the thread class in truss_thread.h) that I use a bool to track the state.  &#60;/p&#62;
&#60;p&#62;As for the redundant code in stop(), the idea it to protect if you call stop() again. pthread does not handle a second cancel or join very well.&#60;/p&#62;
&#60;p&#62;Aldec is a proper simulator.&#60;/p&#62;
&#60;p&#62;It seem to me you are following the code quite well !  I have been coding c++ for 20 years now and would not consider myself an expert.&#60;/p&#62;
&#60;p&#62;I will get back to you when I hear from aldec.&#60;/p&#62;
&#60;p&#62;Take Care,&#60;br /&#62;
mike
&#60;/p&#62;</description>
</item>
<item>
<title>shimrod on "Active HDL (Windows)"</title>
<link>http://forums.trusster.com/topic/active-hdl-windows#post-30</link>
<pubDate>Tue, 04 May 2010 04:10:34 +0000</pubDate>
<dc:creator>shimrod</dc:creator>
<guid isPermaLink="false">30@http://forums.trusster.com/</guid>
<description>&#60;p&#62;Hi &#60;/p&#62;
&#60;p&#62;Im trying to get teal (1.40b) and Truss to work with Active HDL 8.2 under windows XP, with pthreads.2 and the mingw gcc compiler that comes with the simulator.&#60;br /&#62;
I have managed to compile teal (PLI 1) and some of the tests passes. However the mutex_test crashes the simulator. I tried to play with the number of threads and up to 4 works. When I change the no of iterations to 1 I can create many (250) threads. So it seems to me that something with the mutex fails. &#60;/p&#62;
&#60;p&#62;Regarding Truss, I can't compile since the different implementation of pthread_t.&#60;br /&#62;
Truss seems to rely on the id being 0 after thread cancelled?&#60;br /&#62;
Also this seems redundant since teal::stop_Thread is calling both ptrhead_cancel and pthread_join.&#60;/p&#62;
&#60;p&#62;Anyone who has some ideas? Or should I give up and move to linux and a proper simulator?&#60;/p&#62;
&#60;p&#62;Im not an C++ (or C) expert and the code is quite complex to follow.&#60;/p&#62;
&#60;p&#62;Thanks and best regards&#60;br /&#62;
Rolf
&#60;/p&#62;</description>
</item>

</channel>
</rss>

