@ljs@social.kernel.org
@ljs@social.kernel.org avatar

ljs

@ljs@social.kernel.org

Linux kernel mm contributor, kernel/systems developer, writing a book about mm.

C/C++/(rust at some point!)

Book - https://linuxmemory.org/
Me - https://ljs.io/
Music - https://soundcloud.com/distal_music/

Arsenal fan, cat maniac, synth experimentalist. Brit.

Opinions are all my own and represent nobody else.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

Di4na, to random
@Di4na@hachyderm.io avatar

For everyone that calls for ways to make open source more secure, or for all their magical solutions that will provide money and resources to FOSS maintainers, please read this.

This is a rare account of the reality of maintainers, things that are hard, but also how much knowledge and niche expertise you need for anything in there.

That is why just giving money to experts will not help that much. It is too hard to train experts in this. But we may make it easier

http://rhaas.blogspot.com/2024/05/hacking-on-postgresql-is-really-hard.html

ljs,
@ljs@social.kernel.org avatar

@Di4na A very thoughtful piece, another factor I think that is often downplayed is the role of talent - the author of this piece is clearly very talented, but the number of people with enough talent in the world to work on highly complicated software like this is limited, so that adds YET ANOTHER filter on the number of people who can end up working on these things.

I think society as a whole tends to like to act as if anybody if they only want it hard enough could do these things whereas the reality is only a small fraction could.

kernellogger, (edited ) to linux
@kernellogger@fosstodon.org avatar

The 's team just published their thousandth CVE[1]. 🥳 🙃

This happened 78 days after the effort was announced[2].

Note, 26 of the 1003 CVE entries published so far were later rejected. For details check https://git.kernel.org/pub/scm/linux/security/vulns.git/ or https://lore.kernel.org/linux-cve-announce/

[1] https://git.kernel.org/pub/scm/linux/security/vulns.git/commit/?id=55441d0dd1f40c5762cd7cf8c9ca312ed0964c4a

[2] http://www.kroah.com/log/blog/2024/02/13/linux-is-a-cna/

ljs,
@ljs@social.kernel.org avatar

@sima @gregkh @kees @kernellogger what is the benefit in tagging 'could maybe possibly be a security issue' to commits when the definition of that is so broad? Is it really beneficial?

Note that it's not just a 'tag' but for enterprise kernels often entails (potentially significant) work in addressing CVEs?

Perhaps a metadata tag would work better? Potentially-Exploitable or such? :)

Given the controversy around CVEs that might be a lighter touch than the gradual move towards treating more and more commits as if they were known security flaws.

ljs,
@ljs@social.kernel.org avatar

@sima @kees @kernellogger @gregkh @airlied I don't understand, you should backport the entire subsystem constantly? How can that possibly work?

And at that point why are you tagging anything?

Forcing more people into what model?

ljs, to random
@ljs@social.kernel.org avatar

Sometimes I feel like, in an effort for everybody to try to present themselves as hugely empathetic, society has become a hell of a lot less empathetic in reality.

ljs,
@ljs@social.kernel.org avatar

@exa I have suddenly lost all empathy for you with that post

kees, to random
@kees@fosstodon.org avatar

Is anyone using an i386 Linux kernel, built with Clang, and configured with UBSAN? :P Something is going very wrong with the UBSAN handler calls. Anyone wanna help me debug this? https://github.com/KSPP/linux/issues/350

ljs,
@ljs@social.kernel.org avatar

@kees ah yeah @vbabka has had some fun experience with 386 well I think gcov stuff actually wasn't it? But probably similar trashing of registers that shouldn't be trashed.

OK so that's probably irrelevant but sort of ballpark :P

Also 32-bit should die

pony, to random
@pony@blovice.bahnhof.cz avatar

Waiting almost ten minutes in a queue at the post office only to be told it’s a completely different queue but they are closing so maybe I should go there tomorrow. Asking how the fuck am I supposed to tell from me slip, they told me maybe I could try to remember the name of the postman on it as they are separate.

Aaaaargh.

ljs,
@ljs@social.kernel.org avatar

@pony hey @lkundrak is your post as good as this?

pid_eins, to random
@pid_eins@mastodon.social avatar

This is such a bad bad API compat breakage:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e81cd5a983bb35dabd38ee472cf3fea1c63e0f23

It's used all over the place in userspace. In systemd we use it:

  1. to detect if a block device has partition scanning off or on
  2. In our udev test suite, to validate devices are in order
  3. udev rules use it for some feature checks (in older versions of systemd).

And it's even a frickin documented userspace API:

https://www.kernel.org/doc/html/v5.5/block/capability.html

So much about that nonsensical "we don't break userspace" kernel mantra.

ljs,
@ljs@social.kernel.org avatar

@pid_eins "without ever adding a value to an UAPI header"

This not correct? Or did you hardcode?

ljs,
@ljs@social.kernel.org avatar

@pid_eins A header that no longer exists? :)

It's not great that that's referenced, and the doc page in general isn't wonderful (nor was the idea of 'just exposing' something without abstracting in uAPI).

But at the same time, surely relying on you hardcoding kernel-specific value might be a clue that it's not a great interface to use?

So on one hand I agree with you the doc is bad and shouldn't have implied you can do that, but on the other hand a user relying on this should have been more than a little wary.

drewdevault, to random
@drewdevault@fosstodon.org avatar

I do not understand why every modern Unix does not automatically generate a new swap encryption key on boot by default

ljs,
@ljs@social.kernel.org avatar

@drewdevault @levitte @ttmevans obligatory mention of the excellent https://chrisdown.name/2018/01/02/in-defence-of-swap.html article.

The purpose of swap is to balance memory pressure between the page cache and anonymous pages, so when memory pressure occurs the kernel has the ability to evict not-recently-used anon memory instead of just the page cache.

The page cache intentionally keeps a hold of file pages until reclaim so we can avoid disk I/O as much as possible, therefore naturally RAM usage grows over time.

Without swap the kernel has to ALWAYS evict page cache pages meaning that you might end up having to do a bunch of disk I/O that you might not otherwise have to have done.

TL;DR: Swap lets the kernel use memory for what is most in-demand, whether that is anonymous memory or cached file memory. Turning it off means that stale anonymous memory is ALWAYS preferred over cached file memory even if that doesn't make sense.

EDIT: Didn't realise Chris was on fedi - @cdown - incredible article, love it!

ljs, to random
@ljs@social.kernel.org avatar

kernel compile and... (plus now freezes afterwards).

It was working before so it's just strange.

I reapplied thermal paste and reorientated the heat sink and added an extra fan, no dice.

ljs,
@ljs@social.kernel.org avatar

Also getting random temperature spikes constantly. I really don't know how this is possible given I have redone the thermals several times.

I really do not need this

ljs,
@ljs@social.kernel.org avatar

OK weirdly stress-ng with anything set does not cause the problem, only compiling the kernel...!

ljs,
@ljs@social.kernel.org avatar

@kernellogger I have doubts it's really the memory tbh, I see from prior logs it has happened in the past (these mce reports) but it didn't segfault things like it has now...

I am seeing sudden temperature spikes when compiling with all 32 cores (i9-14900k) but no such error when stress-ng'ing.

BTW I have 192 GB of RAM (don't ask me why I can't answer) which is the max the mobo can support so possibly a cause of weirdness... :>)

ljs,
@ljs@social.kernel.org avatar

@kernellogger so if I built with fewer jobs permitted, everything's fine. So this surely is not the memory that's the problem?

Seeing thermals go crazy.

Smells like mobo to me, I wonder if recent kernel (I am on 6.8.2) has made it more possible to hit this somehow? I know the i9-14900k has had some issues

ljs, to random
@ljs@social.kernel.org avatar

So @lkundrak often posts about doors, showing off Brno doors as if they are special, well now I have to show the end game of doors and yes it's British.

Sorry not sorry

ljs, to random
@ljs@social.kernel.org avatar

Since I'm being boring and moaning about the scam de jour again, let me also go 'WTF' about people blindly taking the word of people who stand to make BILLIONS OF DOLLARS on this topic at face value?

Like I don't fucking want to hear 'the CEO of nvidia says AI stuff is incredible', yeah OBVIOUSLY he says that, I mean jesus christ how naive are people??

"I asked the local butcher whether he thinks more people should buy meat and he said they should!"

WOWWWW!!

ljs,
@ljs@social.kernel.org avatar

@kirby @a1ba actually you definitely shouldn't use linux it's terrible

Use OS/2 instead

ljs,
@ljs@social.kernel.org avatar

@a1ba @kirby NO. OS/2.

ljs,
@ljs@social.kernel.org avatar

@lanodan @kirby @a1ba FreeBSD engineers stop crying long enough to give recommendations? Wow

ljs,
@ljs@social.kernel.org avatar

@a1ba @kirby @lanodan yes in the same way you 'share' your living room with a home invader

ljs,
@ljs@social.kernel.org avatar

@kirby @lanodan @a1ba haha no drama from me man, I love these cute hobbyist projects live + let live

ljs,
@ljs@social.kernel.org avatar

@kirby @a1ba @lanodan they have a great license too, like redis, big fan

ljs,
@ljs@social.kernel.org avatar

@kirby @lanodan @mischievoustomato @a1ba they keep their userland tools very up to date too

ljs,
@ljs@social.kernel.org avatar

@lanodan @kirby @a1ba what no BeOS??

Sad times

ljs,
@ljs@social.kernel.org avatar

@lanodan @kirby @a1ba I recommend MacOS to anybody who's not insane enough to use linux, it is my default OS recommendation

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • Leos
  • everett
  • hgfsjryuu7
  • Youngstown
  • Durango
  • slotface
  • rosin
  • InstantRegret
  • ngwrru68w68
  • kavyap
  • osvaldo12
  • thenastyranch
  • DreamBathrooms
  • PowerRangers
  • anitta
  • magazineikmin
  • khanakhh
  • GTA5RPClips
  • mdbf
  • modclub
  • tacticalgear
  • vwfavf
  • ethstaker
  • cisconetworking
  • tester
  • normalnudes
  • cubers
  • provamag3
  • All magazines
    •