Posts
4409
Following
315
Followers
470
Linux kernel hacker and maintainer etc.

OpenPGP: 3AB05486C7752FE1

Jarkko Sakkinen

Edited 4 months ago
I can admit by now: doing this blockchain stuff was not my thing. I tried it for four months and learned my lesson.

Just wanted to put this out loud for the sake of therapy... Now I know a bit of that and in long run that gives me a better perspective of the tech industry.

#lessonslearned
0
0
4
@Aissen i've said some stupid shit about rust and linux, but every time i see a thread like this i feel less embarrassed :-) realizing that i'm not the only person on internet who sometimes put here stupid shit i guess...
0
0
1

Jarkko Sakkinen

a typical rust thread at lkml ;-) can't get over this screenshot:
3
0
3
@vegard god know where all this amazing learning can lead up to ;-) not far away from mars...
0
0
0
@vegard Also, a thing I learned today, URL shorteners at korg: https://korg.docs.kernel.org/git-url-shorteners.html
1
1
1
@vegard https://lore.kernel.org/all/33f3a0a3-a929-4584-af81-6e020671d428@app.fastmail.com/

$ gcc --version
gcc (GCC) 14.2.1 20250110 (Red Hat 14.2.1-7)
Copyright (C) 2024 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Just randomly saw that in my terminal and wanted to put it down somewhere.

https://git.kernel.org/torvalds/c/1e1b6d63d6340
1
0
0

Jarkko Sakkinen

"vmlinux.o: warning: objtool: .export_symbol+0x2e888: data relocation to !ENDBR: stpcpy+0x0"

hmm...

#linux #kernel
1
0
0

Free Software Foundation

is almost here, and we wanted to kick off the celebrations with a little 💗 When was the term "" created?

See if you can answer this correctly before your friends :D

Answer will be posted tomorrow morning.

0% 1973
100% 1986
0% 1994
0% 2001
0
1
0

Jarkko Sakkinen

Edited 4 months ago
It's a non-Rust related principle in kernel that:

1. No code shall pass without a caller.
2. No code shall be more generic than its callers require it to be (because we don't hold API/ABI compatibility per version for in-kernel API's).
0
1
0

Jarkko Sakkinen

A non-rustianated opinion about DMA patches: it should be part of patch set that requires DMA.

I.e. I was eager to try it out but there is no payload to test. I did add lei filter to get everything from rust-for-linux matching DMA so looking forward to do this in future :-)

#rust #linux #kernel #dma
1
1
1

An Open Letter to All European Politicians and Leaders to Abandon X/Twitter:

"By abandoning X/Twitter, leaders can reduce its credibility, promote fairer alternatives, and take a stand against the spread of disinformation, ensuring democratic principles are upheld."

H/T to @everton137 for organizing this - already close to 1,000 signatures:

https://www.openpetition.eu/petition/online/an-open-letter-to-all-european-politicians-and-leaders-to-abandon-x-twitter

1
12
0
@pinkforest I personally buy my music from Bandcamp rather than stream it from Spotify so doing at least in that sense my part :-) And still buy printed books etc.
0
0
0
@josh yeah, it is super convenient
0
0
0
@vitaut good. competition rules. not having death wishes for C++.
0
0
0
I've been for a while planning for a while to add anonymous inodes but every time I start doing that I noticed similar step as the commits below the top-most are. It's better to work like this up until pieces go to right by themselves.

There's also still too much mess to take any bold actions on the use of locks. It also would need much cleaner starting point :-) This is time-consuming, unproductive but also fun...
0
0
0

Jarkko Sakkinen

We meet again after three week break:

https://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git/log/?h=v4l2-loopback

I collect below the top-most commits that are locked in as pretty much dead obvious changes and the top-most commit (whatever its name happens to be) is the dumpster.

I started this by unpurging the top-most just to recall where I was at. Commit ID's live as I might add a new commit in any position (as deep down in history as possible as it makes sense).
1
0
0

so i heard you wanted a filesystem that is copy on write, in-tree (so not zfs), does not eat your data (so not btrfs) and is not maintained by a certain person (so not bcachefs)?

i present you, ocfs2, or oracle's cluster file system version 2

you say that it's for shared network disks in clusters? here i put it as a root filesystem, on a core 2 duo thinkpad, running off a single usb stick on usb 2.0

0
4
0

Jarkko Sakkinen

If I want to test Ubuntu kernel in a VM what do I need to change in this old snippet of mine:

git fetch --no-tags git://git.launchpad.net/~ubuntu-kernel-test/ubuntu/+source/linux/+git/mainline-crack refs/tags/cod/mainline/v6.3-rc7:refs/tags/cod/mainline/v6.3-rc7
git checkout next
git checkout -b next-test
git merge cod/mainline/v6.3-rc7
LANG=C fakeroot debian/rules clean
LANG=C fakeroot debian/rules binary-headers binary-generic

The problem: that URL no longer responds.

#ubuntu
0
0
0
Show older