Linus Torvalds to reject ‘pointless’ pull requests, including AI-generated ones
▼ Summary
– Linus Torvalds announced he will push back on irrelevant pull requests after the release of Linux kernel 7.1 release candidate 5, which he called too large.
– Torvalds criticized the high volume of trivial fixes to random drivers submitted late in the release cycle, arguing they are not worth the churn.
– He noted that many of these late-stage submissions were triggered by AI code review, which he sees as problematic.
– Torvalds stated that non-critical fixes to long-standing issues are inappropriate at this stage, as the focus should be on regressions.
– This is the second week Torvalds has complained about AI complicating kernel development, previously citing unmanageable security list duplication from AI-generated reports.
Linux kernel creator Linus Torvalds has made it clear that he will begin rejecting what he calls “pointless” pull requests, including those generated or reviewed by artificial intelligence. His frustration stems from an increasing number of trivial and poorly timed submissions that, in his view, add unnecessary complexity late in the development cycle.
In his latest weekly state of the kernel update, Torvalds announced the arrival of the fifth release candidate for Linux kernel version 7.1 and did not hide his dissatisfaction. “To the surprise of absolutely nobody by now, rc5 is pretty big. Quite a bit bigger than rc5’s have traditionally been,” he wrote. He acknowledged that most of the new code involves minor driver updates and is therefore less risky, but he questioned whether the churn was justified at this stage.
The typical Linux kernel development process begins with a two-week merge window, during which contributors submit code for the next release. That is followed by up to seven release candidates, each meant to bring the kernel closer to a stable release. By the time rc5 arrives, the expectation is that work is nearly complete. Yet Torvalds noted that many contributions arriving now are “totally trivial stuff” that could have waited.
“These things are ‘fixes’, sure, but at the same time a lot of them are simply so irrelevant that I think they’d be better off in a linux-next tree and get merged during the merge window,” Torvalds wrote. He added that several of these series were “triggered by AI code review.”
Torvalds warned that he will adopt a stricter approach going forward. “So I think I’ll start being a bit more hardnosed about this kind of unnecessary churn this late in the game,” he said. “We are supposed to look for regressions. Non-critical fixes to long-standing issues are simply not appropriate for this late in the release cycle.”
He declared rc5 “too big” and framed his post as “the heads-up that I’ll be pushing back on pointless pull requests with fixes that just aren’t that important.” Torvalds justified the shift by arguing that “these kinds of large rc weeks are not conducive to long-term stability.” He added, “Trivial fixes may be trivial, and have a pretty low chance of causing problems, but ‘low chance’ is still not ‘zero chance.’”
Torvalds ended his update with direct instructions to contributors: “Start looking closer at your pull requests, and ask yourself: ‘Is this really a regression or serious enough that it shouldn’t just go into the development pile?’”
This marks the second consecutive week that Torvalds has voiced concerns about AI’s impact on kernel development. Last week, he complained that “the continued flood of AI reports has basically made the security list almost entirely unmanageable, with enormous duplication due to different people finding the same things with the same tools.”
(Source: Theregister.com)
