<p>First, thank you very much for your reply. </p>
<p>Hmm. I am using ssh -X to access a KVM virtual machine from the host system. Could Ubuntu 11.10 Desktop (my host system) have a different library version than Ubuntu 12.04 Server (my VM)? My VM also doesn't have X installed, but it has various libraries installed due to installing several GUI apps--if lacking a library, would that affect anything and would the benchmarks fail silently?</p>
<p>I will try running the experiments with both the host and VM running Ubuntu 12.04.</p>
<div class="gmail_quote">On Jun 12, 2012 5:01 AM, "Chris Wilson" <<a href="mailto:chris@chris-wilson.co.uk">chris@chris-wilson.co.uk</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Mon, 11 Jun 2012 18:46:13 -0400, Patrick Wilbur <<a href="mailto:patrick.wilbur@gmail.com">patrick.wilbur@gmail.com</a>> wrote:<br>
> Hey all,<br>
><br>
> I'm new to cairo-perf-trace. I'm trying to benchmark ssh -X forwarding<br>
> versus a local X server using the traces in the benchmark/ directory (from<br>
> the traces repository).<br>
><br>
> Over ssh -X, cairo-perf-trace chromium-tabs.trace seemed to work and<br>
> terminate fine. The evolution.trace seems to hang (gave it an hour to<br>
> complete, but it didn't finish). The firefox traces also seem to hang. I'm<br>
> running on Ubuntu 12.04.<br>
><br>
> Any reasons you can think of that this is happening? (Also, any reasons why<br>
> this isn't a good benchmark and, if so, what would be the best way to<br>
> accomplish what I'm trying to test?)<br>
<br>
For the stated purpose of looking at the overhead incurred through SSH X<br>
forwarding, it is fine. Rendering is almost entirely one-way traffic<br>
with very little round-trips so what you are mostly measuring is the effect<br>
of compression upon throughput, and given enough cores shouldn't see a<br>
large difference than rendering natively on the xserver.<br>
<br>
That it explodes timewise is indicative of a bug on your systems,<br>
somewhere between cairo and the remote driver.<br>
-Chris<br>
<br>
--<br>
Chris Wilson, Intel Open Source Technology Centre<br>
</blockquote></div>