<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Fri, Apr 7, 2017 at 10:13 PM, Mark Janes <span dir="ltr"><<a href="mailto:mark.a.janes@intel.com" target="_blank">mark.a.janes@intel.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">José Fonseca <<a href="mailto:jose.r.fonseca@gmail.com">jose.r.fonseca@gmail.com</a>> writes:<br>
<br>
> Hi,<br>
><br>
> Another area of Apitrace I'm struggling with is Android support, because:<br>
><br>
> - I don't have a feeling if / how much it actually matters to users<br>
<br>
</span>FWIW, engineers working on Android at Intel would like to capture traces<br>
from Android to analyze with FrameRetrace. Support for the Android App<br>
store in ChromeOS makes a huge number of titles playable on Mesa, and we<br>
hope to find and fix performance issues with these workloads.<br>
<br>
FrameRetrace needs a lot of work to support retracing on an android<br>
device, but I'm hopeful that the traces can be analyzed on standard<br>
Linux platforms.<br>
<span class=""><br>
> - I have no idea if it even works or not<br>
><br>
> I'm afraid the status quo is unbearable, whereby Android support is<br>
> officially advertised, but users keep complaining issues building/using it,<br>
> and I can't help nor even tell them whether it's supposed to work.<br>
><br>
> Finally, there are alternatives for Android out there. In particular<br>
> Google seems to be actively working on <a href="https://github.com/google/gapid" rel="noreferrer" target="_blank">https://github.com/google/<wbr>gapid</a><br>
><br>
><br>
> So my plan is to:<br>
><br>
> 1) start advertising that Android is no longer supported (still allow the<br>
> build, but place loud warnings)<br>
><br>
> 2) yank out support completely after some time (a month at max)<br>
><br>
><br>
> If by any chance Android still matters and works (or can be easily made to<br>
> work), then I'm counting on people to step up and give a hand to make<br>
> supporting it viable.<br>
><br>
> To halt (or revert) the Android deprecation process I'll need the following:<br>
><br>
> 1) up-to-date instructions on how to build and use with recent Android<br>
> versions / SDKs<br>
><br>
> 2) a Travis-CI build for Android (if somebody helps with item 1 I can help<br>
> with this item 2)<br>
><br>
> 3) a very simple sanity test case using an Android emulator (Hopefully a<br>
> test that can be run from Travis-CI job too, but that can wait. As long as<br>
> any Android user can easily run that test locally and verify things work<br>
> minimally I'm happy.) I don't have experience with Android, nor the<br>
> need/time to learn now, so I'll be relying totally on the community for<br>
> this one too.<br>
<br>
</span>I will ask our engineers if they are willing to help with this work.</blockquote><div><br></div><div>That would be great.</div><div><br></div><div>I'll proceed adding a warning that Android support is in limbo regardless, so that other Android users that aren't subscribed to this list get a heads up.</div><div><br></div><div>Jose</div></div></div></div>