​Linux 4.3 released after Linus Torvalds scraps 'brain damage' code

Linux 4.3 may have been what Linux kernel creator Linus Torvalds describes as a relatively "calm" release but that hasn't prevented him turning the air blue over the quality of a certain piece of networking code.
Written by Liam Tung, Contributing Writer
Linux kernel inventor Linus Torvalds: Networking code in the release candidate triggered an outburst.
Image: Linux Foundation

Linux kernel creator Linus Torvalds has announced Linux 4.3, which he says was a "rather calm" release cycle, even though he had to catch some code that he branded 'stupid' and warned coders to avoid.

Torvalds can take a breather after a demanding schedule in October due to the Linux Kernel Summit in Korea, and some nerves ahead of the general release of Linux 4.3 on Sunday.

"So it 'felt' like the last week of the [release candidate] series was busy, to the point where I got a bit worried about the release. But doing the actual numbers shows that that really was just my subjective feeling, probably due to the kernel summit and travel back home from Korea," Torvalds said in a mailing list.

Still, on the last day of the summit, he found some networking code in the release candidate that triggered an outburst that the Linux kernel inventor has become known for.

"Christ people. This is just ****. The conflict I get is due to stupid new gcc header file ****. But what makes me upset is that the **** is for completely bogus reasons," he noted, honing on new code he said only an incompetent coder would consider legible, efficient, or safe.

"The above code is ****, and it generates **** code. It looks bad, and there's no reason for it. The code could easily have been done with just a single and understandable conditional, and the compiler would actually have generated better code, and the code would look better and more understandable."

He continued with his criticism, arguing that the method used generated bad code and was unnecessary.

"Give me one reason why it was written in that idiotic way with two different conditionals, and a shiny new non-standard function that wants particular compiler support to generate even half-way sane code, and even then generates worse code? A shiny function that we have never ever needed anywhere else, and that is just compiler-masturbation."

"And it's a ****ing bad excuse for that brain damage. I'm sorry, but we don't add idiotic new interfaces like this for idiotic new code like that," he added later.

The developer called out by Torvalds agreed he could have found a simpler way to fix the bug that he was attempting to address but still laid out an argument for why those choices were made. He also didn't appear to take Torvalds' criticisms personally.

The latest outburst follows the departure from the project of a Linux kernel developer, Sarah Sharp, who found the lack of personal respect with the kernel community intolerable.

She had previously criticised Torvalds over verbal abuse on Linux mailing lists. Torvalds has said his style of communication prevents poisoning the community with 'fake politeness'.

As for the final release itself and despite the rant, Torvalds considered Linux 4.3 a "rather calm release cycle until the very end". He noted that changes from the latest release candidate were mostly related to network components.

"We had a network update and a late fix for a x86 vm86 mode bug introduced by the vm86 cleanups, but other than that it's just a collection of various small oneliners all over. OK, the vm86 mode thing was a one-liner too, it was just slightly more nerve-wracking because it looked scarier than it was before people (Andy) figured out what was going on," he wrote.

"And with the release of 4.3, obviously the merge window for 4.4 is open, and let's keep our fingers crossed that that will be an equally calm release," he added.

Read more about Linux

Editorial standards