Change your directory to /sys/devices/system/cpu/intel_pstate

system:/sys/devices/system/cpu/intel_pstate # l
total 0
drwxr-xr-x  2 root root    0 Oct 21 18:45 ./
drwxr-xr-x 14 root root    0 Oct 21 18:45 ../
-rw-r--r--  1 root root 4096 Oct 21 18:45 max_perf_pct
-rw-r--r--  1 root root 4096 Oct 21 18:45 min_perf_pct
-rw-r--r--  1 root root 4096 Oct 21 18:45 no_turbo

# cat max_perf_pct
100

# cat min_perf_pct
32

One of the things that has always been a bit of a struggle in Firefox for Android is getting reliable video decoding for H264. For a couple of years, we’ve been shipping an implementation that went through great heroics in order to use libstagefright directly. While it does work fine in many cases, we consistently get reports of videos not playing, not displayed correctly, or just crashing.

In Android 4.1, Google added the MediaCodec class to the SDK. This provides a blessed interface to the underlying libstagefright API, so presumably it will be far more reliable. This summer, my intern Martin McDonough worked on adding a decoding backend in Firefox for Android that uses this class. I expected him to be able to get something that sort of worked by the end of the internship, but he totally shocked me by having video on the screen inside of two weeks. This included some time spent modifying our JNI bindings generator to work against the Android SDK. You can view Martin’s intern presentation on Air Mozilla.

While the API for MediaCodec seems relatively straightforward, there are several details you need to get right or the whole thing falls apart. Martin constantly ran into problems where it would throw IllegalStateException for seemingly no valid reason. There was no error message or other explanation in the exception. This made development pretty frustrating, but he fought through it. It looks like Google has improved both the documentation and the error handling in the API as of Lollipop, so that’s good to see.

As Martin wrapped up his internship he was working on handling the video frames as output by the decoder. Ideally you would get some kind of sane YUV variation, but this often is not the case. Qualcomm devices frequently output in their own proprietary format, OMX_QCOM_COLOR_FormatYUV420PackedSemiPlanar64x32Tile2m8ka. You’ll notice this doesn’t even appear in the list of possibilities according to MediaCodecInfo.CodecCapabilities. It does, however, appear in the OMX headers, along with a handful of other proprietary formats. Great, so Android has this mostly-nice class to decode video, but you can’t do anything with the output? Yeah. Kinda. It turns out we actually have code to handle this format for B2G, because we run on QC hardware there, so this specific case had a possible solution. But maybe there is a better way?

I know from my work on supporting Flash on Android that we use a SurfaceTexture there to render video layers from the plugin. It worked really well most of the time. We can use that with MediaCodec too. With this output path we don’t ever see the raw data; it goes straight into the Surface attached to the SurfaceTexture. You can then composite it with OpenGL and the crazy format conversions are done by the GPU. Pretty nice! I think handling all the different YUV conversions would’ve been a huge source of pain, so I was happy to eliminate that entire class of bugs. I imagine the GPU conversions are probably faster, too.

There is one problem with this. Sometimes we need to do something with the video other than composite it onto the screen with OpenGL. One common usage is to draw the video into a canvas (either 2D or WebGL). Now we have a problem, because the only way to get stuff out of the SurfaceTexture (and the attached Surface) is to draw it with OpenGL. Initially, my plan to handle this was to ask the compositor to draw this single SurfaceTexture separately into a temporary FBO, read it back, and give me those bits. It worked, but boy was it ugly. There has to be a better way, right? There is, but it’s still not great. SurfaceTexture, as of Jelly Bean, allows you to attach and detach a GL context. Once attached, the updateTexImage() call updates whatever texture you attached. Detaching frees that texture, and makes the SurfaceTexture able to be attached to another texture (or GL context). My idea was to only attach the compositor to the SurfaceTexture while it was drawing it, and detach after. This would leave the SurfaceTexture able to be consumed by another GL context/texture. For doing the readback, we just attach to a context created specifically for this purpose on the main thread, blit the texture to a FBO, read the pixels, detach. Performance is not great, as glReadPixels() always seems to be slow on mobile GPUs, but it works. And it doesn’t involve IPC to the compositor. I had to resort to a little hack to make some of this work well, though. Right now there is no way to create a SurfaceTexture in an initially detached state. You must always pass a texture in the constructor, so I pass 0 and then immediately call detachFromGLContext(). Pretty crappy, but it should be relatively safe. I filed an Android bug to request a no-arg constructor for SurfaceTexture more than two years ago, but nothing has happened. I’m not sure why Google even allows people to file stuff, honestly.

tl;dr: Video decoding should be much better in Firefox for Android as of today’s Nightly if you are on Jelly Bean or higher. Please give it a try, especially if you’ve had problems in the past. Also, file bugs if you have issues!

What a conference!!! openSUSE Asia Summit was unforgettable. The whole event had an amazing feel to it, and I had a rocking time in Beijing. openSUSE really has one of the best and most helpful communities, and the people are amazing. I had the pleasure of interacting with the active organization community. You guys absolutely rock!!!

Here are some of my experiences, and things I learned at the conference:

A little history:
The beginnings of the summit go back a year to Thessaloniki, where the idea of the Asia Summit was first mooted. Due to time constraints and clashes with openSUSE Summit in the US called for the event to be shifted to 2014. I had interacted with Sunny and Max in Thessaloniki and loved the idea of having an openSUSE event in Asia. At oSC14 in Dubrovnik, it was more or less confirmed that the summit would take place.

The Organization Team:
I joined the organization team a little late. The others had already done a lot of the hard work. I helped around a little with the invite letter and the promotion of the logo contest, and trying to find people to help around with the artwork. The opening session, where we welcomed the attendees in our native languages was cool. Also, Sunny’s speech in the beginning, which took us through the past one year was memorable, and I still remember each and every word of it.

I would take this opportunity to thank the openSUSE.Asia Summit
organization team. Today, now the openSUSE.Asia summit has started,
I’m reminded of the journey we took to get here.
I can not forget our weekly meetings, which often lasted to midnight.
I can’t forget 137 cards in trello for the preparation tracking.
And I can’t forget hundreds of emails about the Summit in our mail
boxes.

When we were on the way to reach this summit, we encouraged and
supported each other. Even though we were tired, we never gave up,
because we did believe we would finally be here. It is my honor being
a member of such a great team!

There are 17 people in the organization team, I won’t list everyone’s
name because we are a team, and we couldn’t have make any success
without each of us.

-Sunny

The organization committee did a fantastic job with the event and everything was planned to perfection. I would love to work with you all to host openSUSE Asia Summit next time as well (hopefully in India )

New things:
I absolutely loved the concept of ‘Chops’ where the workshop speakers would put a lovely ‘Geeko’ stamp on the brochure for the participants for the performance in the workshop. More than judging the performance, it gave us a good chance to interact with the attendees and have a lot of fun in the process. The gifts for the speakers and for the chops were great and well thought out. Personally, working with the organization team was very fruitful and I learned how an event of this scale is organized from the ground up. Additionally, being a room coordinator was a novel experience as well.

The Event:
To put the event in a single word: memorable. It was a very well conducted event and the speakers did a great job. The workshops and talks were conducted both the from the point of view of newbies as well as seasoned contributors. I particularly liked Richard’s opening session where the direction that openSUSE (with respect to Factory and Tumbleweed) is taking became clear. There were workshops on Bugzilla and OBS which were really helpful for getting new contributors involved.

Talk is Cheap. Show me the Code:
This was the single biggest lesson I learned during conducting my sessions. While taking the Qt Workshop, I was talking about basic object oriented concepts like Inheritance. The attendees (mostly students from the university) gave me a blank look. I was not sure whether they understood me or not. I ultimately decided to show them some code. They understood that. At that point I realized that there is one universal language that we could communicate with: CODE. That made the job a whole lot easier, and the rest of the sessions went well. I also made some ‘brilliant’ errors during the workshop, which demonstrated some or the other concept with respect to Qt. Overall, had a fun time conducting the workshop.

Food:
Being a vegetarian, I was not sure how I would survive in China. I absolutely indulged myself and tried out plenty of stuff. I have never eaten so much in a conference than I had in Beijing. Thanks to the awesome community guys, specially David who helped me a lot in finding out things to eat. The food was amazing. I can safely conclude that the Chinese take their food seriously. Plus, I learned how to use chopsticks properly. Thanks ftake for that

China and Sightseeing:
This was one trip where I did not do much sightseeing. I had talks for both the days and could not spare the time. To my dismay, I found that visiting the Great Wall requires a full day, and I had just about 6 hours to spare. In the end, I just visited the Forbidden City and the Bird’s Nest (Olympic Stadium). I should have made my travel plans a little better and stayed an extra day.
I found Beijing to be an excellent city. I managed to get around pretty comfortably on the subway despite the language issues. I found the subway system quite effective and very cheap (2 Yuan is dirt cheap). The only problem I faced was the air pollution, which was a little unexpected. Other than that, the people were amazing and really helpful.

TSP:

Thanks to the openSUSE Travel Support program, that I, and many others got to attend the event. It is really an amazing program, and I hope that contributors use it very effectively.

• Refreshed the Update Repositories to contain latest patches
• Applied latest Updates to the Appliance

Following up on KDE’s announcement of the latest stable release, we have now packages available for 12.3 and 13.1 (a 13.2 repository will be made available after it is out). You will find them in the KDE:Current repository. Current users of this repository will get the new release automatically once they update.

Why you should upgrade? You can take a look at the list of changes to get an idea. These fixes touch many important KDE applications, including KMail, Okular and Dolphin.

Packages are also on their way to openSUSE Factory.

As usual, bugs with the packaging should be reported to openSUSE and upstream bugs should be reported to KDE.

Also, if you like what KDE is doing and you feel you can not contribute directly, you may want to support this end of year fundraiser.

Geckos are taking over China, at the openSUSE Asia Summit to be held this coming weekend in Beijing. It is the first time that an openSUSE event of this scale is being held so far east of the Geeko Meridian. The organizing committee has worked their socks off for the event, and things are shaping up well. The schedule is ready and their are some great talks lined up. Overall, it promises to be a great event. As for me, I am going to speak about the Google Summer of Code, openSUSE Activities in India and a workshop on the Qt Framework.So, see you in Beijing

http://summit.opensuse.org

For this task we need newest version of libgreattao and tao-manager application. You can download tao-manager from my home page. You can download newest version of libgreattao from svn of SourceForge,.

Our configuration window

Firstly, we take care about configuration window. In this example we looks on ready class of configuration dialog belonged to “modern” design.

This window contains buttons on the left side, labels and edits. In standard design are placed tabs instead of buttons. In modern design clicking on the button will switch view.

The code are below:

<root>

<variable name="displayed_path" action="change" value="" />

<hbox>

<vbox>

<template path="/category/*">

  <variable name="blank_path" action="change" value="" />

  <if type="equal" variable1="blank_path" variable2="displayed_path">

     <variable name="displayed_path" action="change" function="path" />

     <variable name="active" action="change" function="path" />

     <variable name="active_directory" action="change" value="/" />

      <variable name="active" action="append" variable="active_directory" />

  </if>

  <button dir-for="label">

    <handler>

       <hide variable="active"  />

       <variable name="active" action="change" function="path" />

       <variable name="active_directory" action="change" value="/" />

       <variable name="active" action="append" variable="active_directory" />

       <show variable="active" />

    </handler>

  </button>

</template>

</vbox>

<vbox>

<template path="/category/*/*/text">

<variable name="current_path" action="change" function="path" />

<hide />

<if type="is_children" variable1="displayed_path" variable2="current_path">

  <show />

</if>

<hbox>

<label dir-for="label" />

<edit dir-for="event,input,output" />

<label dir-for="description" />

</hbox>

</template>

</vbox>

</hbox>

<hbox>

<button label="Ok" dir-for="label,event" path="/actions/ok" />

<button label="Cancel" dir-for="label,event" path="/actions/cancel" />

<button label="Save" dir-for="label,event" path="/actions/save" />

</hbox>

</root>

We can see one handler, few variables, conditions and show/hide elements. Show/hide elements are changing visibility:

1. For parent element,, if there’s no “name” and “variable” attribute
2. For elements with given path, if there’s “variable” attribute
3. For elements with given name, if there’s “name” attribute

What we can do to show/hide elements once button is clicked? After all we must select method(using path or name). We select path and controling elements using children path “/category/*/, because buttons are present in “/category/*” and elemetns to control are in path “/catrgory/*/*”. In first step we hide all elements using “hide” element without attributes for template:

<template path="/category/*/*/text">

This way will hide all elements without category buttons. We must also show firstly created category. This code will complete this task:

<if type="is_children" variable1="displayed_path" variable2="current_path">

   <show />

 </if>

Condition typw “is_children” checks that path are child of other path. In case we need it touches only children of matched element(without parent), we must end value with slash. Jn this case we check if “current_path” are child of “displayed_path”. “current_path” is set above and the value of it is cuirrent path. We set also “active variable”, which points to children fof active element(we add slash at the end, using”append”). Before setting value of variable “displayed_path” we checks if that variable are not s,lash(this variable will contains slash near after view is loaded).

In next step we take care about showing/hidding elements on button’s click event. Our handler will complete this task. In first step, we hide active element using “active” variable. We must remember that firstly created element will be active. In next step we set “active” variable using clicked button’s action path. We also append slash to it. In  next step we show elements belonged to clicked button. “active” variable will contain path of clicked element, so after clicked another button displayed elements(without buttons) will be hide.

This is the end of describing configuration window. There’s only one matter. Once application delete active elements, there’s no elements visible(buttons are only visible). Other situation is with tabs. There’s framework will display another tab automatically.

---

Custom file management window for tao-manager

Next example are about adding file information dialog for my file manager. We will change view of file manager windolw, but only for this one application. Firstly, we need to create directory “$HOME/.tao”. In this directory we need to create directory “$HOME/.tao/tao-manager” and “$HOME/.tao/tao-manager/mWidgets” and “$HOME/.tao/tao-manager/mWidgets/app”. In next step we need to create file with UI paths definitions(“$HOME/.tao/app”). In this directory we will create UI defintion directory for my file manager:

tao-manger = tao-manger

We must now copy configuration file for mWidgets design:

cp /usr/share/tao-design/mWidgets/config ~/.tao/tao-manager/mWidgets

We must now create file called “file manager” in “$HOME/.tao/tao-manager/mWidgets/app” directory. This file should looks like below:

<root>

<menu>

<template path="/actions/*">

<submenu dir-for="label">

<attr-connect name="label" function="last-dir" />

<template path="*">

<item dir-for="label,event">

<icon dir-for="icon" width="8" height="8" />

<attr-connect name="label" function="last-dir" />

</item>

</template>

</submenu>

</template>

</menu>

<hbox>

<vbox>

<template path="/fileviews/*/list/items/*">

<variable name="name" action="change" function="path" />

<variable name="name" action="append" value="/name" />

<attr-connect name="widget-name" variable="name" />

<template path="data/*">

<hbox>

  <label>

   <attr-connect name="label" function="last-dir" />'

  </label>

  <label dir-for="label" />

</hbox>

</template>

<hide />

</template>

</vbox>

<template path="/fileviews/*">

<vbox>

<template path="path_edit">

<edit dir-for="input,output,event" />

</template>

<list>

<template path="list/columns/*">

  <column>

    <attr-connect name="name" function="last-dir" />

    <attr-connect name="label" function="last-dir" />

  </column>

</template>

<template path="list/items/*">

  <selection>

    <template path="selected">

       <handler>

          <hide name="path" />

          <variable name="path" action="change" function="remove_last_dir" />

          <variable name="path" action="append" value="/name" />

          <variable name="call-path" action="change" function="path" />

          <show name="path" />

          <call var="call-path" />

       </handler>

    </template>

    <template path="data/*">

       <data dir-for="label">

         <attr-connect name="column-name" function="last-dir" />

         <attr-connect name="label" function="last-dir" />

       </data>

    </template>

  </selection>

</template>

</list>

</vbox>

</template>

</hbox>

</root>

Our information panel are definied in template with path pattern “/fileviews/*/list/items/*. In this step I must mentoin about widgets names. Many widgets can have the same name and one widget can have many names. Name are used to control visibility of widget, but in future it will be used in other tasks. We can define widget’s name usaing “attr-connect” tag with “name” attribute equal to “widget-name” and giving variable name. Our name is actions’s path with appented string “/name”. Why we don’t use action’s paths? We are using the same path for elements in lists, so using name could causing list’s elements to show/hide. I must mentoin, that displaying/hidding elements of lists are not implemented, but maybe it will be implemented in future.

Described tempalte contains template with relative path pattern “data/*”. In this pattern exist label with text equal to last directory of path to this action. There’s also label connected to value of current attribute of active file. All attributes of file are hidden by default.

Take also look at only one handler in this file. It exist in template with relative path “selected”, “selected” is event invoked, when a file is selected. “path” variable point to active file. It is empty by default, so nothink will be hidden.  This handler first will hide element(vbox for file attribute). In next step it will get path without last element(“/selected”) and next append value “/name” for it. This is way of generation path of element to show. “call-path” variable contains path pointed to “selected” action. This variable is used to call event connected with this action,.

There’s only one problem. Names of file attributes in left panel are not set by application.

 We use names obtained from path of template(last directory). It can be solved in very simplem way. You must only replace first code bellow with second code below:

1)

<hbox>
<label>
<attr-connect name=”label” function=”last-dir” />
</label>
<label dir-for=”label” />
</hbox>

2)

<hbox>
<variable name=”column-path” action=”change” value=”/fileviews/” />
<variable name=”column-path” action=”append” function=”nth-dir” param=”2″ />
<variable name=”column-path” action=”append” value=”/list/columns/” />
<variable name=”column-path” action=”append” function=”last-dir” />
<label>
<attr-connect name=”label” function=”last-dir” />
<attr-connect name=”label” function=”label” param=”column-path” />’
</label>
<label dir-for=”label” />
</hbox>