| Post Reply | RetroFed Forum Forum Index » Linux |
| Author | Message | ||
|---|---|---|---|
Ek-Hou-Van-Braai
piefed.social
Ek-Hou-Van-Braai@piefed.socialJoined: 2025-05-14 |
|
||
| 14 Original | |||
0xKesh
lemmy.dbzer0.com
0xKesh@lemmy.dbzer0.comJoined: 2026-06-17 |
Shit site keeps kicking me out for my ad-blocker |
||
| 15 | |||
L3ft_F13ld!
lemmy.dbzer0.com
reseller_pledge609@lemmy.dbzer0.comJoined: 2024-11-07 |
Shit site keeps kicking me out for my ad-blocker
Full articleAfter a few revisions of his position on AI-assisted contributions to the Linux kernel, Linus Torvalds has posted his own first LLM-assisted patch. The development branch purports to fix an old issue with massive cache contention when many CPUs are reading the same file — a scenario potentially encountered when using shared libraries, databases, and web servers. Several developers had already made attempts over the years at fixing the issue, all with varied results, none quite ideal as an all-encompassing solution. Torvalds posted his own draft for solving the problem in early 2024, and has used it on his own laptop since that time. However, he recently took the plunge and decided to "try to actually solve this thing using an LLM." He's now posted his solution in a development branch, and claims that he's seen a 23x performance improvement in a dedicated synthetic benchmark, though he notes that actual real-world impact will be far less significant, as "no real load" will witness those figures. If we had to guess, this improvement would benefit server workloads far more than desktop machines. Over time, Torvalds has broadly taken a pro-AI stance on LLM tooling, at least within the purview of the tightly controlled, review-heavy environment of the Linux kernel code. Last July, he stated he's "willing to absolutely put [his] foot down" on AI haters, pointing out that an automated code review tool the kernel team uses outperforms humans, and that it caught basic programming mistakes that shouldn't have been submitted in the first place. The team is even changing its policy to no longer require that patch authors disclose they used an AI tool when generating their code. Torvalds once again took an assertive stance, reiterating that users should treat LLMs as yet another tool, likening it to compilers or editors. In his view, he simply expects that "people do their best and [...] understand what they are doing." Some might say that few people in the world are better-suited to properly employ an LLM than Torvalds, and by extension, kernel maintainers. There's an entire universe of difference between a senior master programmer using the LLM to speed up the boring parts of his work, and some newbie with a credit card going "claude make linux faster plz no mistakes." However, It's not all roses. Thanks to how adept AI agents have become at recognizing common security flaw patterns, the Linux security mailing list became "unmanageable" earlier in the year. Linux has been fixing CVEs at an astonishingly rapid pace of around 2,000 per release. Still, the team's hand is being forced into dropping older hardware drivers, as they're often the target of numerous security reports, despite their negligible install base. |
||
| 26 | |||
macniel
feddit.org
DmMacniel@feddit.orgJoined: 2024-06-22 |
Yes we get it, he drank the Kool-AId... |
||
| 18 | |||
0xKesh
lemmy.dbzer0.com
0xKesh@lemmy.dbzer0.comJoined: 2026-06-17 |
::: spoiler Full article After a few revisions of his position on AI-assisted contributions to the Linux kernel, Linus Torvalds has posted his own first LLM-assisted patch. The development branch purports to fix an old issue with massive cache contention when many CPUs are reading the same file — a scenario potentially encountered when using shared libraries, databases, and web servers. Several...
Much appreciated! |
||
| 9 | |||
|
HelloRoot
lemy.lol
HelloRoot@lemy.lol Joined: 2025-02-15 |
Is this the one where the LLM generated code introduced some severe bug (memory leak or buffer overflow or something), then Linus went on vacation and another dude cleaned it up? https://lkml.iu.edu/2610.1/13698.html afaik If yes, then it annoys me that the articles on this are so biased. Because it actually shows the middle ground reality of using LLMs in development instead of the utopia/dystopia extreme views we keep seing online. |
||
| 63 | |||
macniel
feddit.org
DmMacniel@feddit.orgJoined: 2024-06-22 |
Is this the one where the LLM generated code introduced some severe bug (memory leak or buffer overflow or something), then Linus went on vacation and another dude cleaned it up? https://lkml.iu.edu/2610.1/13698.html afaik If yes, then it annoys me that the articles on this are so biased. Because it actually shows the middle ground reality of using LLMs in development instead of the utopia/dyst...
Racing condition between threads. But yeah I believe its this one. |
||
| 29 | |||
L3ft_F13ld!
lemmy.dbzer0.com
reseller_pledge609@lemmy.dbzer0.comJoined: 2024-11-07 |
Much appreciated!
Happy to help. If a site is gonna be absolute garbage then they don't deserve traffic. |
||
| 12 | |||
Ek-Hou-Van-Braai
piefed.social
Ek-Hou-Van-Braai@piefed.socialJoined: 2025-05-14 |
Is this the one where the LLM generated code introduced some severe bug (memory leak or buffer overflow or something), then Linus went on vacation and another dude cleaned it up? https://lkml.iu.edu/2610.1/13698.html afaik If yes, then it annoys me that the articles on this are so biased. Because it actually shows the middle ground reality of using LLMs in development instead of the utopia/dyst...
Did AI generate that? |
||
| 2 | |||
|
HelloRoot
lemy.lol
HelloRoot@lemy.lol Joined: 2025-02-15 |
Did AI generate that?
? |
||
| 5 | |||
Ek-Hou-Van-Braai
piefed.social
Ek-Hou-Van-Braai@piefed.socialJoined: 2025-05-14 |
?
Did AI generate that race condition commit of linus? |
||
| 4 | |||
|
HelloRoot
lemy.lol
HelloRoot@lemy.lol Joined: 2025-02-15 |
Did AI generate that race condition commit of linus?
(sorry, I thought by "that" you meant my comment or the content of the link or something else) As far as I understand, Linus did an LLM assisted coding session that resulted in code with a race condition. But to what extend he or the LLM is specifically responsibe for the bug, we don't know because he didn't share the session, only the result. |
||
| 20 | |||
Ek-Hou-Van-Braai
piefed.social
Ek-Hou-Van-Braai@piefed.socialJoined: 2025-05-14 |
(sorry, I thought by "that" you meant my comment or the content of the link or something else) As far as I understand, Linus did an LLM assisted coding session that resulted in code with a race condition. But to what extend he or the LLM is specifically responsibe for the bug, we don't know because he didn't share the session, only the result.
Thanks, I thought he only now started using LLM's in commits to write code, based on the article. But I haven't really followed him closely |
||
| 6 | |||
katy ✨
piefed.blahaj.zone
cupcakezealot@piefed.blahaj.zoneJoined: 2025-06-06 |
then distros need to start rejecting any linus contributed code and fork the kernel. |
||
| 4 | |||
Ek-Hou-Van-Braai
piefed.social
Ek-Hou-Van-Braai@piefed.socialJoined: 2025-05-14 |
then distros need to start rejecting any linus contributed code and fork the kernel.
Just what the Linux community needs, more fragmentation |
||
| 18 | |||
|
onlinepersona
programming.dev
onlinepersona@programming.dev Joined: 2023-09-01 |
Man, this is going to garner a lot of hate by people who hate AI just by virtue of it being AI. |
||
| 3 | |||
|
refalo
programming.dev
refalo@programming.dev Joined: 2024-04-13 |
then distros need to start rejecting any linus contributed code and fork the kernel.
lmao |
||
| 6 | |||
|
crandall
lemmy.today
crandall@lemmy.today Joined: 2026-08-11 |
Man, this is going to garner a lot of hate by people who hate AI just by virtue of it being AI.
Yes that's us |
||
| 11 | |||
|
Denys Nykula
piefed.social
nykula@piefed.social Joined: 2026-02-04 |
then distros need to start rejecting any linus contributed code and fork the kernel.
A fork would need lots of initiative from existing kernel developers. Distro people are more of userspace integrators, less system programmers. I think distros will experiment with rebasing against different kernels, maybe a revival of Debian GNU/kFreeBSD or something. At the end of the day though, it's up to laptop and phone manufacturers whether to put money into developing hardware support on top of those. |
||
| 4 | |||
BJW
lemmus.org
Epp@lemmus.orgJoined: 2026-02-12 |
Man, this is going to garner a lot of hate by people who hate AI just by virtue of it being AI.
Welcome to Lemmy, where the hive mind has decided progress is bad, and will criticize anyone who disagrees until they either shut up or become one with the hive mind. People who still speak their own disagreement are the minority, and are rewarded with copious downvotes. They've made up their mind, and discourse is heavily discouraged. |
||
| 1 | |||
|
YaBoyMax
programming.dev
YaBoyMax@programming.dev Joined: 2023-06-09 |
then distros need to start rejecting any linus contributed code and fork the kernel.
I genuinely can't tell if this is sarcasm or not and that makes me uneasy. |
||
| 8 | |||
|
bjaress
ttrpg.network
bjaress@ttrpg.network Joined: 2026-06-28 |
Is this the one where the LLM generated code introduced some severe bug (memory leak or buffer overflow or something), then Linus went on vacation and another dude cleaned it up? https://lkml.iu.edu/2610.1/13698.html afaik If yes, then it annoys me that the articles on this are so biased. Because it actually shows the middle ground reality of using LLMs in development instead of the utopia/dyst...
It's almost like the person going on vacation comes away with a positive view, and the person cleaning up after them comes away with a negative view . . . |
||
| 15 | |||
|
ikidd
lemmy.dbzer0.com
ikidd@lemmy.dbzer0.com Joined: 2026-04-23 |
I genuinely can't tell if this is sarcasm or not and that makes me uneasy.
The idiocy is self-explanatory. |
||
| 3 | |||
artyom
piefed.social
artyom@piefed.socialJoined: 2025-06-08 |
Man, this is going to garner a lot of hate by people who hate AI just by virtue of it being AI.
Nobody hates AI 'by virtue of being AI", there are a dozen different legitimate problems. |
||
| 3 | |||
artyom
piefed.social
artyom@piefed.socialJoined: 2025-06-08 |
then distros need to start rejecting any linus contributed code and fork the kernel.
Honestly you'd have to fork an entire distro because at this point shameless AI development is taking place from the bottom to the top. Debian, KDE, kernel, etc. |
||
| 2 | |||
|
MangoCats
feddit.it
MangoCats@feddit.it Joined: 2025-02-05 |
Did AI generate that race condition commit of linus?
Maybe Linus saw the race condition but just fixed it badly, again. https://dirtycow.ninja/ "This is an ancient bug that was actually attempted to be fixed once (badly) by me 11 years ago in commit 4ceb5db9757a... but that was then undone due to problems on s390 by [another] commit." |
||
| 2 | |||
apotheotic (she/her)
beehaw.org
apotheotic@beehaw.orgJoined: 2023-07-11 |
Just what the Linux community needs, more fragmentation
Open source when someone forks something for a reason that aligns with my worldview: "The beauty of open source! Anyone can fix a problem!" Open source when someone forks something for a reason that doesn't align with my worldview: |
||
| 5 | |||
|
MangoCats
feddit.it
MangoCats@feddit.it Joined: 2025-02-05 |
(sorry, I thought by "that" you meant my comment or the content of the link or something else) As far as I understand, Linus did an LLM assisted coding session that resulted in code with a race condition. But to what extend he or the LLM is specifically responsibe for the bug, we don't know because he didn't share the session, only the result.
I believe Linus has clearly stated his position that he is specifically responsible for the bug, regardless of whether he uses LLM agents, Ouija boards, or vim to create it. |
||
| 5 | |||
|
Avicenna
programming.dev
Avicenna@programming.dev Joined: 2025-11-07 |
Welcome to Lemmy, where the hive mind has decided progress is bad, and will criticize anyone who disagrees until they either shut up or become one with the hive mind. People who still speak their own disagreement are the minority, and are rewarded with copious downvotes. They've made up their mind, and discourse is heavily discouraged.
You can't just put every anti AI sentiment into the "progress is bad" bin, if you do that is the same type of behaviour as a hive mind. Many people have legitimate concerns about AI but equally many conflate AI with billionaires who exploit resources (to develope and sell AI products) for their own gain. In general, most people conflict with progress because progress is greatly in the monopoly of billionaires who have no qualms for using those progresses for exploitation. |
||
| 2 | |||
| Post Reply |
Back to top |