Hi Chris,<br><br>Yes, I do recall the debacle. But, as you say, that should have been worked around a long time ago already.<br><br>It would be interesting to know if anyone else has noticed anything similar? <br>For the moment I'm content running just one of the cores.. ;)<br>
<br>/Robert<br><br><br><div class="gmail_quote">2009/2/6 Chris Cannam <span dir="ltr"><<a href="mailto:cannam@all-day-breakfast.com">cannam@all-day-breakfast.com</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d">On Fri, Feb 6, 2009 at 8:18 PM, Robert Jonsson <<a href="mailto:spamatica@gmail.com">spamatica@gmail.com</a>> wrote:<br>
> Thinking back I realize that I stumbled on this problem on the last rebuild<br>
> some years ago but ended up running single core without investigating why<br>
> this was.<br>
<br>
</div>The symptom sounds rather like the infamous old AMD64 non-monotonic<br>
clock resulting from clocks not being properly synchronised across<br>
cores. But I had thought that workarounds in the kernel, JACK, or<br>
both had made this a non-issue for most systems installed in the last<br>
couple of years. (I am quite sure that a workaround dedicated to<br>
solving specifically this problem was merged to JACK trunk quite a<br>
number of releases ago, and any contemporary JACK release will have<br>
it.)<br>
<br>
This page: <a href="http://www.sound-man.co.uk/linuxaudio/kernel.html" target="_blank">http://www.sound-man.co.uk/linuxaudio/kernel.html</a> refers to<br>
the problem.<br>
<br>
I realise I'm not offering any solution here, just pointing out that<br>
your problem sounds the same as this one.<br>
<font color="#888888"><br>
<br>
Chris<br>
</font></blockquote></div><br>