The father of Linux is angry again in the New Year! Angrily denounced Google kernel contributors: I will not accept this kind of garbage code

Compilation丨Noah

Produced by 51CTO Technology Stack (WeChat ID: blog51cto)

Linus Torvalds recently published a highly critical post on the Linux kernel mailing list for Google's contributors, making scathing views on filesystem recommendations. The controversy centers on the inode, which according to Red Hat is a unique identifier for metadata on a specific file system.

Inode issues have been a hot topic on the recent Linux kernel mailing list, sparking many discussions. In particular, Google kernel contributors Steven Rostedt and Torvalds had a heated exchange on this issue. In the discussion about "making all files and directories have the same inode number", it was pointed out that inodes may still have some role when using tar for file archiving. However, Torvalds believes that the era of inodes has passed, triggering more discussion and controversy. This shows that the Linux community has different views and positions on the inode issue, which also reflects the complexity and importance of the issue.

He wrote: "Yes, the inode number was once important and had its own unique historical background. However, we should now move towards a more advanced and efficient file system. The inode number is no longer unique descriptor, we are in an era of constant change.”

Despite this, the debate about inodes continues. Eventually Rostedt suggested that all inodes should have unique numbers.

The wording and tone of Torvalds's response have been unusual in recent years.

In 2018, Linus Torvalds announced that he would temporarily leave and receive help for emotional control. He has apologized for previously making flippantly offensive remarks in emails to other Linux programmers and project contributors, acknowledging that his actions were unprofessional and unjustified. He said he was deeply sorry for the attacks, especially when they were taken personally.

Torvalds expressed contrition mainly because the Linux kernel mailing list is actually the workplace of many contributors. Abuse of verbal violence is not only detrimental to creating a pleasant work environment, but also affects the participation of volunteers and maintainers. Considering that Linux projects rely on these contributors, a poor working environment can have a negative impact on the entire project.

In response to Rostedt's suggestion about unique inode numbers, Torvalds first said: "Don't make things more complicated than they need to be."

Then his words became more intense:

"Oh my god, stop copying VFS layer functions! It was a bad idea last time, and it's an extremely bad idea this time. I won't accept this kind of garbage code."

Torvalds' main criticism of Rostedt was that he believed the Google developer did not fully understand the subject matter involved - something Rostedt later acknowledged.

However, before that, Torvalds had already launched his own verbal "attack." As stated below:

You copied a function without understanding why it did a specific thing, so your code is complete garbage.

Emphasis on this again.

For the next period, the debate continued in a relatively calm tone. Torvalds made suggestions that he believed would better address Rostedt's concerns.

The father of Linux wrote that he does not intend to tackle the problem immediately because he has already wasted enough time on it and he is too far behind on other responsibilities to invest in it now. energy.

In response to Torvalds’ statement, Rostedt responded with a rather targeted observation:

I completely understand your current work load and schedule. I really appreciate you taking the time to discuss this topic and provide valuable insights and alternatives. I will consider your suggestion carefully and reexamine my implementation. In order to ensure the quality of our code and the overall stability and efficiency of the Linux kernel, I will delve deeper into the underlying mechanisms related to the file system to avoid such misunderstandings and invalid work from happening again. Thank you again for your guidance. I will adjust my development direction accordingly and strive to achieve better solutions.

Ironically, one of the responsibilities I've been putting off is fixing eventfs while writing a support document on how maintainers deal with work stress and burnout. :-p

On Sunday, Torvalds began to publish news on the happy topic of the Linux kernel: Linux kernel 6.8-rc2 version was officially released.

He pointed out that included in rc1 is an "amdgpu scheduling bug that can cause the desktop to hang (it will eventually recover, but the recovery time may be long enough for most users to choose to restart the computer rather than wait)."

"This issue seems to affect a fair number of people."

Fortunately, a btrfs bug that Torvalds himself encountered did not appear in rc1 because it was noticed before release.

He wrote in the post: "Regardless, I hope that with the launch of rc2, we are now entering a more stable stage in the release cycle, and those types of issues that may affect a large number of testers have been resolved. So , I hope that subsequent repairs will be more detailed and try not to affect common core configurations."

"So, please feel free to test it out. It's safe now. You trust me, right?"

Of course, if the question is only about developing the kernel itself.

After Linus Torvalds released the Linux kernel 6.8-rc2 version, he encouraged developers and testers to actively participate in testing the new version. Although challenges such as bug fixing and feature optimization will inevitably be encountered during the kernel development process, the Linux kernel can be continuously improved and enhanced through continuous iterative updates and cooperation and feedback among community members. Therefore, for those who trust and care about kernel development, participating in testing not only helps improve the quality of the kernel, but is also one of the key steps to push the entire open source ecosystem forward.

Reference link: https://www.theregister.com/2024/01/29/linux_6_8_rc2/?td=rt-4a

The above is the detailed content of The father of Linux is angry again in the New Year! Angrily denounced Google kernel contributors: I will not accept this kind of garbage code. For more information, please follow other related articles on the PHP Chinese website!