<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Sat, Aug 15, 2015 at 12:32 AM, Alberto Aguirre <span dir="ltr"><<a href="mailto:alberto.aguirre@canonical.com" target="_blank">alberto.aguirre@canonical.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">That will also be the result of the scaling governor. I was surprised too that arale (Meizu) only has an interactive governor available.<div>It's quite hard to measure performance consistently and effectively with such a governor (specially for comparison purposes). </div><span class=""><div><br></div><div>><span style="font-size:12.8000001907349px">I'm more concerned about how can we keep </span><span style="font-size:12.8000001907349px;background-color:rgb(255,255,255)">phone</span><span style="font-size:12.8000001907349px"> graphics performing as well as they do during touches, even when we're not touching them?</span></div></span><div><span style="font-size:12.8000001907349px">I don't think there's escaping the fact that we will need to tune the governor to match the Ubuntu workload. </span></div><div><span style="font-size:12.8000001907349px"><br></span></div><div><span style="font-size:12.8000001907349px"><br></span></div><div><span style="font-size:12.8000001907349px"><br></span></div></div><div class="gmail_extra"><br><div class="gmail_quote"><span class="">On Fri, Aug 14, 2015 at 3:55 AM, Daniel van Vugt <span dir="ltr"><<a href="mailto:daniel.van.vugt@canonical.com" target="_blank">daniel.van.vugt@canonical.com</a>></span> wrote:<br></span><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">It's not just frequency either. On arale (Meizu) for example, smoothness correlates directly with whether multiple CPU cores are online or not:<br>
<br>
cat /sys/devices/system/cpu/online<br>
<br>
Usually the kernel only keeps one core online, which makes Unity8 stutter. But if you touch it enough then the second core (out of eight) comes online and everything is smooth. I wonder if more aggressive use of threads might help...</span><div><div><span class=""><br>
<br>
<br>
On 14/08/15 16:48, Daniel van Vugt wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi all,<br>
<br>
In testing performance optimisations on various phones, I keep running<br>
into an annoying hurdle.<br>
<br>
Although you can optimise your Mir server/clients in such a way that<br>
they're smoother more often, there's an additional variable outside of<br>
Mir and Unity that gets in the way. That seems to be frequency scaling<br>
done by the kernel. Sometimes on desktops too, but I'm mostly concerned<br>
about phones here.<br>
<br>
I find it suspicious that on some devices you can turn stuttering into<br>
smoothness just but touching the screen a lot. But the smoothness soon<br>
goes away when you're not touching the screen. In the extreme case, if<br>
you're logged into the phone remotely you will also notice the system<br>
can become unusably slow when the screen has turned off. That's useful<br>
for a real phone's battery life, but it serves to illustrate that the<br>
kernel is doing a lot behind the scenes. I'm more concerned about how<br>
can we keep phone graphics performing as well as they do during touches,<br>
even when we're not touching them?<br>
<br></blockquote></span></div></div></blockquote></div></div></blockquote><div><br><div>I'm not sure these are what you want to, but if you're testing on Arale, you can enable cpus by<br></div><div> /proc/hps/num_base_perf_serv [# of little] [# of bigs] <br></div><div> ex) sudo /proc/hps/num_base_perf_serv 2 2 (2 littles and 2 bigs)<br></div><div>and also can set gpu with the max freq by <br></div> echo 0 > /sys/module/pvrsrvkm/parameters/gpu_dvfs_enable<br></div><div>(though I'm recommending this only for testing.)<br></div><div></div><div></div><div></div><div></div><div></div><div></div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div><span class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
- Daniel<br>
<br>
</blockquote>
<br>
-- <br></span>
Unitymirteam mailing list<br>
<a href="mailto:Unitymirteam@lists.canonical.com" target="_blank">Unitymirteam@lists.canonical.com</a><br>
Modify settings or unsubscribe at: <a href="https://lists.canonical.com/mailman/listinfo/unitymirteam" rel="noreferrer" target="_blank">https://lists.canonical.com/mailman/listinfo/unitymirteam</a><br>
</div></div></blockquote></div><br></div>
<br>--<br>
Mir-devel mailing list<br>
<a href="mailto:Mir-devel@lists.ubuntu.com">Mir-devel@lists.ubuntu.com</a><br>
Modify settings or unsubscribe at: <a href="https://lists.ubuntu.com/mailman/listinfo/mir-devel" rel="noreferrer" target="_blank">https://lists.ubuntu.com/mailman/listinfo/mir-devel</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature"><div dir="ltr">Chunsang(Paul) Jeong</div></div>
</div></div>