Linux 6.17 RISC-V Updates Rejected ⁢by Linus Torvalds – Late Submission and “Garbage” Code cited

Linus ⁣Torvalds, the creator ‍of Linux, has rejected a recent submission of RISC-V code updates for inclusion in the upcoming Linux 6.17 kernel release. The rejection, delivered with ⁣characteristic bluntness, stems from both ⁢the ‍late arrival of the pull request and the inclusion of code torvalds deemed “garbage.”

Late Submission and Travel Concerns

The RISC-V updates were submitted on Friday, nearing the end of the Linux 6.17 merge window, ‍which is slated to close Sunday‍ with the release ⁢of the 6.17-rc1 candidate. torvalds explicitly requested early submissions due to his international travel schedule this week, a request ⁣the RISC-V maintainers failed to meet.

However, the timing wasn’t the sole issue.Torvalds made it‍ clear that even an on-time submission wouldn’t have guaranteed acceptance given the quality of the code presented.He stated he expected a “good” pull request if an⁤ early submission wasn’t possible.

“garbage” Code and Generic Header Pollution

The core of Torvalds’ criticism centered around code he described as “garbage” being added to generic header files. This ‍isn’t a matter of ⁢functionality, but of code‍ hygiene and maintainability. He specifically called⁤ out a newly introduced “helper” function,makeu32fromtwou16(),as a prime exmaple of problematic code.

“That thing makes the world actively a worse place to live,” Torvalds wrote in his rejection message. He argued that using⁤ the helper function obscures the code’s intent, making it less readable and understandable than a straightforward expression.He emphasized that the function’s placement in a generic header file meant it could potentially pollute other projects and encourage ⁣the adoption of similarly unhelpful code.Torvalds’ concern isn’t simply about aesthetics.He believes such ‍additions actively degrade the codebase, making‍ it ⁤harder for developers to understand and maintain the kernel. He stressed⁤ the⁣ importance⁤ of keeping generic header files clean and free of specialized, potentially detrimental “helpers.”

A ‍Strong warning and Path Forward

Torvalds issued a firm ⁢warning to the ⁣submitters: no more late pull requests, and no more non-RISC-V ⁣specific code polluting ⁣generic headers. He indicated that while he hopes the core RISC-V code⁢ itself is sound, the obligation for⁢ ensuring its quality rests ⁣with the maintainers.

The rejected code will have⁤ to ⁢be resubmitted for consideration in Linux 6.18, but only if it arrives early in the merge window and is free of the⁣ “garbage” that prompted the rejection.

Conciliatory Response from RISC-V Maintainers

The submitters of the patches responded with a conciliatory ⁤message, acknowledging the criticism. They pledged to improve their submission process and focus on code quality. “I’ll ‍stop being late, and hopefully that helps with the quality issues,” they stated.

This incident serves as a reminder of Linus ‍torvalds’ uncompromising commitment to code quality and his willingness to‍ enforce strict⁤ standards for⁣ kernel advancement. It also highlights the importance of⁤ adhering to established submission guidelines, especially when the project lead is facing logistical challenges.