Forgot your password?
Close
typodupeerror
AI Linux

AI Reviews Bring 'New Normal' to Linux Release Candidates: Lots of Bug Fixes (phoronix.com) 65

Posted by EditorDavid from the hardening-at-work dept.
Linux Torvalds expects Linux 7.2 should be released next weekend "unless something really bad pops up," Torvalds said while announcing today's release candidate.

But there's something interesting about Linux 7.2-rc7, writes Phoronix. "By the time the Linux kernel typically hits a -rc7 release things have usually settled quite well. But in today's world of AI/LLM coding/review agents, the kernel activity continues at an all-time high." Tons of bug fixes continued to trickle in across the kernel spectrum for all sorts of issues. The HWMON hardware monitoring subsystem saw several critical and high severity bug fixes, on the memory management side was a nasty race condition leading to a use-after-free in the kernel for the past eight years, Btrfs restored its fixup worker infrastructure to deal with silent data loss, lots of AI patches in the networking realm, and the kernel was patched for the Safe RET Interrupt Vulnerability.

Linus Torvalds wrote in the 7.2-rc7 announcement:


"Another week, another -rc.

I can't say that I'm exactly thrilled about the size of this all, but it is what it is: the new normal with a lot of fixes, many of them due to review by various AI tools.

And nothing looks particularly scary per se — it's just that there's a lot here. Most of it is fairly small, although we have a couple of larger diffs: s390/zcrypt fixes stand out in the diffstat, and so does btrfs bringing back the fixup worker infrastructure. And some netfilter ipset fixes.

But aside from a few places like that, most of this is just lots of tiny fixes. It's pretty much spread all over — drivers (gpu, sound, networking, you name it), filesystems, core networking, arch code...

AI Reviews Bring 'New Normal' to Linux Release Candidates: Lots of Bug Fixes More | Reply

AI Reviews Bring 'New Normal' to Linux Release Candidates: Lots of Bug Fixes

Comments Filter:
  • I much prefer a big RC7 and an extended RC phase, than use-after-free race conditions and silent data loss. Thank you AI.

  • Linux Torvalds expects Linux 7.2 should be released next weekend

    So when will AI take over the world? One of the most recognizable names in tech history, still active. Does anyone who posted or at Slashdot review? Of course not. Could AI fix this problem? Probably so.

    • Re: (Score:2)

      by jd ( 1658 )

      You see, Linus has embraced transhumanism and is testing out the new kernel as a supplementary brain function, using MOSIX to offload all of the irritable comment generation at yet more nonsense on the mailing list to an Elizabot that he has written specifically to do this. This saves his actual brain for real work.

  • I mistakenly applied it to my win7 era Manjaro box and was left with an unresponsive black screen, I can login to a terminal CTL-ALT-F2, etc but I don't know how to fix it. I guess I shouldn't put it into my new box and let it try to adjust to it before I put it back in the old box and let it go back to its old configuration, I probably corrupted the install. I can't imaging what possessed me to try the release candidate when the 6 level linux was a safe option and the stable 7 level was also available.

    I t

    • I, too, often say to myself "bit rusty on the old command line, maybe I should hand roll an unreleased kernel".

      • Lol no, I just ticked the wrong box in the package manager when given the option. I din't hand roll nottin.
        now, if I was more familiar with pacman (is it?) I could probably roll it back but maybe first I shou;d turn off the quick boot in the bios so I can select a different line in grub or turn off auto login. or something. I'm not sure how worthwhile it is to salvage an old box when age is diminishing my brain power. I can just slap the drive back into the new box and take advantage of the extra terabyte.

        • Re: (Score:3)

          by DMFNR ( 1986182 )
          If your Grub timeout is very short or "0", meaning you don't see a Grub screen when you boot (common on single install systems), when you restart start tapping Esc when you see your computer logo and it should bring up the Grub menu. If it doesn't work try turning off Fastboot, usb keyboards especially might not be initialized in time for the Esc thing to work.

          Your old Kernel should be there in Grub waiting for you so you can just boot in to that one and figure it out from there. I don't know Pacman, but

          • Thank you for this confirmation of what I was thinking. I mostly don't use that box, it is like third on the chain on my desk behind this new one using the same OS and a mac mini I use to do my taxes all attached to the same monitor.
            BTRFS seems a bit more finicky as to how you attach extra drives to it, but I can't say I have explored the available information accessible from the net. I've used LINUX for quite some time going back to Mandrake but was never particularly an elite user.

            Anyways, not particularl

  • Are why under the new artificial intelligence regime, we *can* have nice things.

  • I'm going to go out on a limb here (Score:3)

    by rsilvergun ( 571051 ) on Sunday August 09, 2026 @09:44PM (#66281186)
    And say that if you've had a bug in your code for 8 years and the code is actually in production and being used then the bug wasn't all that important.

    AI seems to be finding a lot of bugs that people just weren't bothering to find because why would you? I have certainly come across bugs in my own code that don't really hurt anything or that require such very specific and uncommon circumstances that I'm hardly surprised I didn't fix them sooner. Yeah a machine can and we'll find them but so what?

    If it didn't cost anything to fix them that would be fine but I don't know about you but my power bill just went up 15% and despite the lies they're telling me it's painfully obvious that it's the fund data center power usage. Also Amazon is putting a data center online that is powered completely by pollution belching gas turbines that are going to crank out more greenhouse gases then literally anything in the entire country. Climate change is real and it's causing droughts. Those droughts are driving up the price of beef and chicken.

    So how much is an LLM fixing esoteric bugs Worth to you? In a few years you might be eating bugs because of it...

    • Re: (Score:1)

      by evanh ( 627108 )

      They are new bugs in new AI generated code.

      • Why do you speak when you have no idea what you are talking about. Again, at least RTFSummary.

        • Re: (Score:2)

          by evanh ( 627108 )

          Take a guess at what a Release Candidate (RC) is about.

          • I have written kernel device drivers and have been a Linux developer and user since 1997. I've built multiple custom Linux distributions used in embedded systems for various companies over the years. Since you have literally no idea how Linux development works, you should probably find out who you are talking to before making ridiculous snarky comments.

            • Re: (Score:2)

              by evanh ( 627108 )

              Just because you aren't using AI to write code doesn't mean others haven't done so. Linus is on record saying AI is welcome as long as it works. Well, what works is still up for grabs. People will be using the new tools. And slop is one thing that AI is already well known for.

              • There is a reason why Linus, who is much smarter than you, says what he says and does what he does. Nothing about kernel development has changed. There is not now, nor has there ever been, a way to determine who wrote the code. The code lives and dies on its merits and gets reviewed. It neither gets accepted or rejected because it is AI code. Code is code. There is an entire system of development at play of which you have literally no understanding. I suggest you learn how to read an article summary before

                • Re: (Score:2)

                  by evanh ( 627108 )

                  Not really a rebuttal there. AI slop is submitted and there is more bugs because of it.

                  • A claim you are making not only without any evidence to substantiate it, but in direct contradiction to the evidence that your claim is untrue. If it were true then Linus would not be allowing the use of AI. To claim otherwise is absurd. Some people aren't smart enough to figure out that they aren't smart and you have proved to be one of those people. Off you go now ...
    • The fact that you think that people "aren't bothering" to find bugs because "why would you?" says everything anyone with a clue needs to know about your "expertise" in software development.

    • Re: (Score:2)

      by AmiMoJo ( 196126 )

      A lot of these bugs are security related. Even if there isn't an obvious path to exploitation right now, being in the kernel means that they might be useful as part of a chain later on.

      • It's not just that there isn't an obvious path it's that there isn't any practical path. That's why nobody is finding these bugs. And what I'm saying is is we need to think about how much water and electricity we're pissing through and how much damage all this nonsense is going to do to our lives and our economy in exchange for finding a handful of bugs that could have sat there till the end of time without anybody giving a shit.

        This is not a free endeavor. We are in the middle of a massive drought and

        • Re: (Score:2)

          by AmiMoJo ( 196126 )

          Other people are going to be using the same tools to find the same bugs. We really have to do it because Linux is critical infrastructure.

    • Re: (Score:2)

      by twdorris ( 29395 )

      So how much is an LLM fixing esoteric bugs Worth to you?

      First, who said these are all esoteric bugs? If it fixes a few esoteric bugs but also fixes several significant bugs, then you need to define that gradient where you consider it worthwhile because that's the situation we're in. It's ignorant to claim these tools are being used by thousands and thousands of *competent* coders to do nothing more than fix trivial typos.

      Second, you know what kind of bugs suddenly show up out of nowhere to wreak havoc on the world? The dormant bugs that have "been in use" for

      • The ones that weren't completely esoteric we're already being found by people. All this is doing is saving them a little bit of time and companies aren't going to repurpose those programmers they're going to fire their asses and pocket the savings.

        So in exchange for a ridiculous amounts of water and electricity we're getting something that has no value to speak of. Not that anyone around here wants to hear that because hey cool new tech right? The idea of an antisocial technology is something that nerds

    • The issue is there is a large class of bugs that were never patched out, they were just hidden because they were slightly beyond the human attention horizon. Now that AI has started to improve, that horizon is effectively machine-extended, so bugs that sat unnoticed for 8 years are suddenly surfacing. It isn't that they weren't important, it's that the bug was effectively secured through obscurity, and now that obscurity is gone. Therefore, we need to patch the bugs the AI can spot, because an attacker usin

      • It isn't that they weren't important, it's that the bug was effectively secured through obscurity, and now that obscurity is gone.

        It wasn't. That was never a thing. It was just a bug that a human hadn't bothered to find yet. It wasn't obscured, it was only unexplored.

        • An insecure server is the street equivalent of parking a car with the window rolled down and valuables in sight. Someone *will* steal your crap if given a clear enough opportunity. The catch is, actually finding and exploiting bugs requires a high level of talent, time, and dedication, so it mostly suffices to make the bugs that do exist sufficiently hard to exploit. That is, at its most fundamental level, security through obscurity. The problem is, AI changes the talent and time issue.

    • Esoteric bugs have a way of becoming security nightmares as people more fully understand them. *shrug*

      Everything should be picture perfect and coherent all of the time. This is the real world and that is impossible... but striving for it is not wasted effort.

  • I get it. I do. I've seen "AI slop". You know who pushes out AI slop, though? Sloppy ass coders. That's who.

    All you people that think AI is a fad that always produces crap need to do a serious introspection. Think, review, and decide if it's REALLY the tool or the user. Because I can tell you with 100% certainty that when *I* use it, *I* get clean, functional, sometimes beautiful, well designed, and well implemented results.

    Used properly with the right scope and proper mindset, the current generation

    • "Because I can tell you with 100% certainty that when *I* use it, *I* get clean, functional, sometimes beautiful, well designed, and well implemented results."

      That's what everyone thinks. Thinking that doesn't make you special.

      • Re: (Score:2)

        by twdorris ( 29395 )

        That's what everyone thinks. Thinking that doesn't make you special.

        Excellent insight. Thanks for your foundational contribution to this dialogue. Now run along, the adults are talking.

        • Now run along, the adults are talking.

          When you find one, ask them if they can find your sippy cup and put you down for a nap.

    • Re: (Score:1)

      by wed128 ( 722152 )

      Sloppy ass coders. That's who.

      The vast majority of coders are sloppy ass coders. Every software team I've ever worked on is propped up by the 20% of people who actually give a shit. Making the other 80% more productive is a net negative.

    • Re: (Score:2, Insightful)

      by pz ( 113803 )

      Used properly with the right scope and proper mindset, the current generation of AI tools are incredibly effective and they keep getting better and better.

      This is the idea that the anti-AI folks are getting wrong. I'm a vastly more efficient and better writer, of code or of English prose, with an assistant than without. Assistants are, however, imperfect, whether human or artificial.

      The current generation of LLM assistants are equivalent in quality of output to having a junior human associate, but at 1/100th the cost. And they're about 100x faster. Is their output perfect? No, but neither is the work-product of a human associate. Does the LLM (or human)

      • I had an argument just last week about didactic presentation with a stubbornly anti-AI professor for whom I have been hired as an instructor for the course he runs. I made the assertion that very shortly, the expectation will be that anyone in that profession who does not use AI will not be bringing their full potential to the task.

        The current generation of LLM assistants are equivalent in quality of output to having a junior human associate, but at 1/100th the cost.

        So you think a junior associate is the equivalent to a chat bot that costs two orders of magnitude less than actually paying a junior associate?

        At the same time I presume you don't consider yourself the equivalent of a chat bot and do not expect your income to be reduced two orders of magnitude?

        Do you expect all of the professors students to be hired directly into senior positions? If not who is going to hire them when they can hire 100 chat bots for the price of one junior?

        Personally, I think it is already true that anyone who is fundamentally a communicator of ideas and is not using AI is not achieving their best

        This reads suspiciously similar

        • Re: (Score:2)

          by pz ( 113803 )

          So you think a junior associate is the equivalent to a chat bot that costs two orders of magnitude less than actually paying a junior associate?

          You're making a classic mistake of reversing an association: "this A is a B" does not mean "every B is an A".

          Which is to say, you're ignoring the fact that junior associates have attributes and capabilities that LLMs lack, in a wince-inducing oversight. Indulge me with a Sophoclean interrogatory: what can humans do that LLMs cannot?

          • You're making a classic mistake of reversing an association: "this A is a B" does not mean "every B is an A". Which is to say, you're ignoring the fact that junior associates have attributes and capabilities that LLMs lack, in a wince-inducing oversight.

            I've done nothing except ask questions. This was your statement not mine: "The current generation of LLM assistants are equivalent in quality of output to having a junior human associate, but at 1/100th the cost"

            I'm simply asking obvious follow up questions. What would those attributes be? Is "quality of output" the work product or were you referring to something else? If it was the work product which I presume you were referring to what of substance do you believe remains? If you meant something else

Slashdot Top Deals

The brain is a wonderful organ; it starts working the moment you get up in the morning, and does not stop until you get to work.

Close