CSS Attribute Selector madness?

Today I stumbled upon a CSS attribute selector which made me think.

div.fooClass[class]
{
height: 29px;
}

The interesting thing here is the [class] part of the CSS. If I understand correctly, this means, that elements should be selected which have the class attribute. I understand why someone might be interested to style all images with alt-tags differently: img[alt]. But is there any legitimate use of [class]? Cause this only says: select elements which have a class assigned… It does not even check WHICH class, it selects just all classes. And it gets weirder when it is combined with a specific class like fooClass in this example. Has anybody an idea what this means?

Lenovo T460(s) suspend problem while on battery

In this blog post I want to summarize my investigations about problems concerning suspend with my lenovo t460 notebook. (Note: This problem seems to also exists with the lenovo T460s notebooks. More information about debian on the t460 can be found here.)

(The informations here are verified with a debian testing installation with kernel version 4.4. The problems is reported by people using other distributions too.)

What is the problem and how can you reproduce it?

• If the device is on the AC, closing the lid causes the notebook to suspend. This works perfectly – in other words: The notebook wakes up after opening the lid and is usable.
• If the notebook is on battery, closing the lid causes the notebook to freeze. In other words: If I open the lid again, everything is frozen. Not only the UI but I’m also not able to change to TTY1, 2, ..
• Interestingly enough, manually suspending with systemctl suspend works without a glitch.

I reported this problem on the debian forums and the thinkpad subreddit.

There is an entry in the kernel bug tracker Bug 113551 – intel_pstate=no_hwp else Thinkpad T460s freezes on lid close on battery power. The bug entry indicates that the problem is also to be found kernel versions up to 4.5.0-rc6. It also looks like the source of the problem is found, patches are being tested. I also filed a debian bug entry for this problem as was suggested to my when asking in the debian IRC channel.

So it seems that newer kernel versions won’t have this problem. There is a workaround till this happens. You can add the intel_pstate=no_hwp parameter to the GRUB_CMDLINE_LINUX_DEFAULT section of the /etc/default/grub file. Attention: Don’t forget to run update-grub after every change to this file. Also: This may cause your notebook to use more power – the battery may empty itself faster.

I will try to update this blog post after more information, changes in the status of this problem occur. I maybe even try to patch the debian kernel with a patch and test if this helps to fix the suspend problem.

Update 2016-04-28: After upgrading to the Debian 4.5.1-1 (2016-04-14) x86_64 GNU/Linux kernel version, the suspend problem is gone. I can close the lid while being on batteries and the notebook is going into suspend and leaves it correctly.

Some interesting thoughts about Functional Programming

In this InfoQ.com video presentation called “Functional Programming You Already Know” Kevlin Henney is trying to reveal functional programming pattern where you would not expect it (i.e. in excel as the world most popular used function programming language). The presentation starts a little bit slow but it’s worth your time.

The end of the Iceweasel Age?

One of my most visited blogposts is the one explaining why there exists something called iceweasel the browser and summarizing the reasons why debian renamed firefox. Now there has been some new activities which I don’t want to hide from you.

For roughly the past decade, Debian has shipped the Mozilla desktop applications (Firefox, Thunderbird, and Seamonkey) in a rebranded form that replaces the original, trademarked names and logos with alternatives (Iceweasel, Icedove, and Iceape). Originally, this effort was undertaken to work around incompatibilities between the Debian Free Software Guidelines (DFSG), the Mozilla trademark-usage policy, and the licenses of the Mozilla logos. But times—and policy wordings—change, and Debian now seems poised to resume calling its packages by the original, upstream Mozilla names.
(Source: lwn.net)

So it seems that the iceweasel package some of you know and love may be gone in the future.

Lenovo T460 and Debian Gnu/Linux

Today I got my new notebook. A Lenovo T460 Thinkpad. In this blogpost I want do document my problems or absence of problems with installing and using Debian Gnu/Linux. I paid around 930 Euro via the u:book program (where you can get notebooks cheaper if you are a student/pupil or working at a university). The version I bought included 8GB Ram and 256 GB SSD.

tldr: I’m writing this under Debian Testing (stretch) with wireless activated and working audio. Also the screen is working with the expected 1920×1080

More detailed information about the hardware (lspci). You can also look at the the output of lspci -v

00:00.0 Host bridge: Intel Corporation Sky Lake Host Bridge/DRAM Registers (rev 08)
00:02.0 VGA compatible controller: Intel Corporation Sky Lake Integrated Graphics (rev 07)
00:14.0 USB controller: Intel Corporation Device 9d2f (rev 21)
00:14.2 Signal processing controller: Intel Corporation Device 9d31 (rev 21)
00:16.0 Communication controller: Intel Corporation Device 9d3a (rev 21)
00:17.0 SATA controller: Intel Corporation Device 9d03 (rev 21)
00:1c.0 PCI bridge: Intel Corporation Device 9d10 (rev f1)
00:1c.2 PCI bridge: Intel Corporation Device 9d12 (rev f1)
00:1f.0 ISA bridge: Intel Corporation Device 9d48 (rev 21)
00:1f.2 Memory controller: Intel Corporation Device 9d21 (rev 21)
00:1f.3 Audio device: Intel Corporation Device 9d70 (rev 21)
00:1f.4 SMBus: Intel Corporation Device 9d23 (rev 21)
00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection I219-V (rev 21)
02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. Device 522a (rev 01)
04:00.0 Network controller: Intel Corporation Wireless 8260 (rev 3a)

Installation

First I tried to use the Debian testing netinstall image. But at that time (2.3.2016) the installer was broken because of some dependency problem between systemd and the ifupdown package. (I suppose this bug ticket is concerned with the problem.) My motivation to use the testing version was to have the better hardware support which was important for me because of the novelty of the T460. My fears now where that the Debian Stable Version (at this time: Jessie) would not be able to install itself on the notebook. But interestingly enough: It did work. The sound, wireless, touchpad and screen had some problems. But this was no problem, because after the first reboot I edited the /etc/apt/sources.list and replaced the jessie keyword with testing. After that used apt-get upgrade and apt-get dist-upgrade upgraded me to the testing version.