Google engineer's RISC-V code criticized as 'garbage' by Linux founder Linus Torvalds, who also accused it of worsening the world; delays in submissions by the dev also called out
Linus Torvalds, the creator and lead developer of Linux, has publicly rejected a RISC-V code contribution from Google engineer Palmer Dabbelt for inclusion in the Linux 6.17 kernel. The rejection was due to poor code quality and late submission, as Torvalds labelled the contribution as "garbage."
In a sharp but reasoned response, Torvalds pointed out several issues with the code. The most significant problem was the addition of a "useless" helper function for combining two unsigned 16-bit integers into a 32-bit integer, which was placed in generic header files that could affect other architectures. This change, Torvalds argued, confuses the codebase and is unacceptable, regardless of timing.
Torvalds also criticised changes to generic header files that were not RISC-V specific, which could negatively impact other parts of the kernel. He emphasised that such changes make the world actively a worse place to live and advised no further late pull requests or "garbage outside the RISC-V tree."
The community's reaction to Torvalds' directness has been mixed. Some see it as a reflection of his commitment to quality, while others find it challenging to maintain constructive collaboration. Torvalds' response, however, was backed by technical reasoning and examples, reflecting the Linux kernel community's high standards on code quality and maintainable development timelines.
The RISC-V code contribution was sent as a pull request for inclusion in the Linux 6.17 kernel, but it was too late in the merge window, violating the Linux kernel development process that demands punctual contributions for review and integration. Torvalds warned the developer responsible, stating, "You're on notice: no more late pull requests, and no more garbage outside the RISC-V tree."
Despite the rejection, Torvalds urged the engineer to prepare better, cleaner patches early for Linux 6.18. He emphasised the importance of early submissions to ensure a thorough review and integration process.
In a separate development, the phrase "random turd files" was used in a headline due to Torvalds' criticism of the RISC-V code contribution. However, it's important to note that Torvalds did not personally attack anyone involved in the contribution.
For up-to-date news, analysis, and reviews, users can follow Tom's Hardware on Google News or sign up for Tom's Hardware's newsletter. The Italian proverb "Il mele catta più mosche, che non fa l'aceto" suggests that charm can be more effective than harshness, but in the world of Linux kernel development, Torvalds' unfiltered and direct approach seems to be an efficient way to communicate high standards and expectations.
- Despite the rejection of his RISC-V code contribution, the Google engineer was encouraged by Linus Torvalds to prepare better, cleaner patches for the upcoming Linux 6.18 kernel, emphasizing the importance of early submissions for a thorough review and integration process.
- linus torvalds, in his rejection of the RISC-V code contribution, pointed out that the addition of a "useless" helper function in generic header files, which could affect other architectures, confuses the codebase and is unacceptable, regardless of timing.
