The 3.12 kernel is out | Hacker News Reader

The 3.12 kernel is out

lwn.net

106 pointsjeltz39 comments

I am always suprised that people are not concerned about this:

Linus @ https://lwn.net/Articles/572706/

"But the fact that I'm going to be (effectively) off-line next week means that I'm not opening the merge window for 3.13 yet - since I won't have the bandwidth to really do merges anyway. That doesn't mean that you can't send me pull request for the merge window early, of course - maintainers can always send their pull requests early rather than late, if they have everything lined up and ready. But if you have some feature that still wants polishing, you basically get a free week to do that. So the two-week merge window for 3.13 will start a week from now. You have an extra week. But that also means that I will be doubly disappointed in anybody who then leaves their merge request until the end of that two-week merge window."

Linus is in absolute control of the kernel development -- this works quite well, but if he gets sick or whatever then how quickly and well will the community find a substitute?

There's a good discussion of 'what if Linus gets hit by a bus?' in the comments of http://lwn.net/Articles/393694/ [On the scalability of Linus].

Don't forget the segfault.org classic: "What If Linus Torvalds Gets Hit By A Bus?" - An Empirical Study

http://segfault.org/writing/segfault.org/Bus.html

From what I saw and of the other core committers could take over his job. If Linus is not merging things, commits are probably just piling up on the repos of the respective subsystem responsible persons.

How many people download and compile sources directly from kernel.org?

i do. whats the big deal about that?

It's a good day for me and many others who are suffering from bug https://bugzilla.kernel.org/show_bug.cgi?id=59841 , black screen after GRUB.

I hope they do a bug-fix only release. There's been an issue with Intel Wireless cards since 3.8. Had to downgrade to 3.7 to get consistent Wifi...

3.11 made a huge improvement for my wireless. Have you tried that?

Yes. It works well for a while, then randomly crashes.

Is it the iwl4965 driver? For me using 11n_disable=1 makes it work more reliably.

iwl3945 actually. Trying your fix though, so far so good. On 3.11.6 right now...

If you have an easy way to reproduce that could you report a bug about that upstream? The only way to reproduce it for me is when using a hotel WiFi, which gives a quite narrow window to report bugs/test patches.

There are >4000 known bugs in the kernel. They would be probably busy fixing just the significant ones until "4.0" when just doing bugfixing.

The easiest thing to do is to report to the bugtracker of your distro and see.

He better not miss out on 3.14.

Pi wouldn't count on it.

But can we say that with any real precision?

Let's circle back to this.

These puns are starting to feel repetitive.

It's a real circlejerk.

FOR PETE'S SAKE GUYS THIS ISN'T REDDIT!

and now the no humour brigade begins to circle the wagons.

3.12 allows for vga reset. So the unresponsive, bind/unbind passed trough video cards are a thing of the past.

For the people that use it could be a big deal.

(Tested on arch, 3.12-RC6, latest QEMU)

Thanks for pointing that out! Good to know I won't need that work around in the future. I need to get back into this, got a couple of graphics cards lying dormant in a server just begging to be passed through to guests!

Check the later pages in that topic - there were few quirks if you use intel IDG, but I mostly managed to do it painless. https://bbs.archlinux.org/viewtopic.php?id=162768

Wait, what? So VirtualBox can now get exclusive access to a graphics card for mostly painless 3d acceleration in VMs? That's pretty cool.

Would definitely be very supportive of a bugfix release.

Is there something significant in 3.12?

It says right there in the post:

* improvements to the dynamic tick code,

* support infrastructure for DRM render nodes,

* TSO sizing and the FQ scheduler in the network layer,

* support for user namespaces in the XFS filesystem,

* multithreaded RAID5 in the MD subsystem,

* offline data deduplication in the Btrfs filesystem, and more.

In case anyone else is curious, DRM in the kernel stands for Direct Rendering Manager [1]. It has nothing to do with the DRM that's usually discussed on HN.

[1] https://en.wikipedia.org/wiki/Direct_Rendering_Manager

Considering the negative connotations of DRM, I am surprised it isn't named something else.

The name Direct Rendering Manager has been around since the '90s, since the XFree86 days. So I'm pretty sure this abbreviation predates that of the "bad" DRM.

Good to know. Thanks.

All TLA are overloaded.

According to Phoronix benchmarks some workloads benefit greatly from the changes in CPU frequency scaling.

http://www.phoronix.com/scan.php?page=article&item=linux_312...

Progress marches on, but as much as I enjoy some good kernel porn, only one in a thousand HN readers will be affected in any way at all.

Significant for me, they reverted this bug - https://bugzilla.kernel.org/show_bug.cgi?id=61781

Could not get a dbus connection following resume from suspend. Very annoying.

Macbook Air (2013) support works better in 3.12.

Looks like it fixes a bug in eDP clamping that scrambles by screen.

Reply on news.ycombinator.com