<br><br><div class="gmail_quote">On 19 September 2011 13:20, James Courtier-Dutton <span dir="ltr"><<a href="mailto:james.dutton@gmail.com">james.dutton@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">On 19 September 2011 12:25, Tim Draper <<a href="mailto:veehexx@gmail.com">veehexx@gmail.com</a>> wrote:<br>
><br>
> this is one of the things i am planning; it would solve a load of 'issues'<br>
> (be it fan noise, or driver conflicts). it would require re-wiring the sat<br>
> cables, and also a more powerful server (im thinking an upcoming virtualbox<br>
> install that can handle PCI(e) passthru).<br>
><br>
<br>
</div>I am a Linux kernel developer.<br>
PCI(e) passthru is a minefield, or at least it was rather problematic<br>
a year ago.<br>
The issue is the physical address, register mapping for DMA transfers.<br></blockquote><div><br>ah, thats interesting. i had a quick look at XEN hypervisor, seems todo the PCI passthru stuff. I've never experimented with it though. <br>
<br></div></div>