<div dir="ltr"><div>I have used boost atomic or just a boost mutex with a scoped_lock when I needed to protect, It implements cross platform the best avail on a platform. <br><br></div><div><br><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Mar 17, 2015 at 9:07 AM, Robert Smith <span dir="ltr"><<a href="mailto:smith@cognitec.com" target="_blank">smith@cognitec.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
Ok, I will implement my own TaskScheduler class, but operations that
read-modify-write memory are not atomic with multiple
processors/cores. I guess the large portion of users these days will
be using multicore x86 processors.<br>
<br>
I asked this question on stackoverflow and the responses support the
research I have done myself.<br>
<br>
<a href="http://stackoverflow.com/questions/29099094/is-the-c-operator-atomic-with-a-multicore-processor" target="_blank">http://stackoverflow.com/questions/29099094/is-the-c-operator-atomic-with-a-multicore-processor</a><br>
<br>
My updated application with a protected triggerEvent() method
didn't cure the problem, one (out of six) of the streams stopped
after an hour. I then modified BasicTaskScheduler to protect all
reads/writes of the fTriggersAwaitingHandling and it has been
running for over 24 hours now. Time will tell but it looks
promising.<br>
<br>
If I'm to speculate why this hasn't been a problem before I would
guess that the large majority of users are not using the triggers at
runtime, e.g, they are using the file sources. And for those who are
using a live source, if they only have one stream then the task
schedulers' SingleStep() method is probably waiting in the select()
call when the triggerEvent() method is called making it unlikely
that they conflict.<br>
<br>
Also each thread in my application has it's own event trigger id.<br>
<br>
I'm not even suggesting that you change the implementation but at
least be aware of the problem and document it for future users.<br>
<br>
<br>
<div>On 03/16/2015 05:50 PM, Ross Finlayson
wrote:<br>
</div>
<blockquote type="cite">
<div>
<blockquote type="cite">
<div>
<div bgcolor="#FFFFFF" text="#000000">Having
thought about it some more, I don't think it's enough to
just protect calls to triggerEvent() because
'fTriggersAwaitingHandling' is still modified by
BasicTaskScheduler::SingleStep().<br>
</div>
</div>
</blockquote>
<div><br>
</div>
The intention of the code that implements “BasicTaskScheduler”
was that:</div>
<div>1/ The event loop code (which is always just a single thread,
remember) does nothing with “fTriggersAwaitingHandling” except
clear (using “&=~”) a single bit that was already set, and</div>
<div>2/ The other thread(s) that can trigger event(s) do nothing
with “fTriggersAwaitingHandling” except set (using “|=“) bit(s)
that were not already set.</div>
<div><br>
</div>
<div>If these two conditions are met, *and* if the ‘clear a bit’
and ‘set a bit’ operations are atomic, then the code will be
thread-safe.</div>
<div><br>
</div>
<div>(If, OTOH, you’re worried that the ‘clear a bit’ and ‘set a
bit’ operations might not be atomic on your system, then you’d
need to write your own “TaskScheduler” subclass instead.)</div>
<div><br>
</div>
<div>Looking at the code again just now, however, I realized that
the event loop code does not always satisfy condition 1/. Line
183 of “BasicTaskScheduler.cpp” should be changed from:</div>
<div><span style="white-space:pre-wrap"> </span>fTriggersAwaitingHandling
= 0;</div>
<div>to</div>
<div><span style="white-space:pre-wrap"> </span>fTriggersAwaitingHandling
&=~ fLastUsedTriggerMask;</div>
<div><br>
</div>
<div>I have just released a new version (2015.03.16) of the code
that fixes this. You should upgrade to this version.</div>
<div><br>
</div>
<div><br>
</div>
<div>You should also make sure that your own code satisfies
condition 2/. I.e., you should make sure that your code is
never calling “triggerEvent()” more than once with the same
‘event trigger’ value, unless you’re sure that the first event
(for that event trigger) has already been handled. (In
particular, you should never call “triggerEvent()” with the same
‘event trigger’ value from more than one thread.)</div>
<br>
<div>
<span style="border-collapse:separate;color:rgb(0,0,0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-align:-webkit-auto;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px"><span style="border-collapse:separate;color:rgb(0,0,0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-align:-webkit-auto;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px">Ross Finlayson<br>
Live Networks, Inc.<br>
<a href="http://www.live555.com/" target="_blank">http://www.live555.com/</a></span></span>
</div>
<div><br>
</div>
<br>
<br>
<fieldset></fieldset>
<br>
<pre>_______________________________________________
live-devel mailing list
<a href="mailto:live-devel@lists.live555.com" target="_blank">live-devel@lists.live555.com</a>
<a href="http://lists.live555.com/mailman/listinfo/live-devel" target="_blank">http://lists.live555.com/mailman/listinfo/live-devel</a>
</pre>
</blockquote>
<br>
</div>
<br>_______________________________________________<br>
live-devel mailing list<br>
<a href="mailto:live-devel@lists.live555.com">live-devel@lists.live555.com</a><br>
<a href="http://lists.live555.com/mailman/listinfo/live-devel" target="_blank">http://lists.live555.com/mailman/listinfo/live-devel</a><br>
<br></blockquote></div><br></div>