Jarkko Sakkinen
Posts
4943Following
327Followers
492OpenPGP: 3AB05486C7752FE1
Jarkko Sakkinen
jarkko
Is there generic probe style timeout in Linux kernel that you could pair with eBPF script? I.e. can you say somehow that "if a process context is stuck more than X seconds, please trigger this eBPF". I.e. then you could inspect location of PC, dump stack and whatnot.
1
0
1
Jarkko Sakkinen
jarkko
@ncopa Most issues I've had with RISC-V have been way or another related to CPU caches. The ISA spec defines these properties with "lowest common denominator" approach, which is a design decision, as the spec's goal is not to define CPU architecture but more like seed for bunch of architectures. That said, I don't believe tho that this could be the root cause for such a wide spread :-)
E.g. when you read or write memory you don't necessarily always have any silver bullet to know how the hardware caches an address range, and often synchronization can require some extra dance with memory fences depending on which RISC-V CPU you run it.
E.g. when you read or write memory you don't necessarily always have any silver bullet to know how the hardware caches an address range, and often synchronization can require some extra dance with memory fences depending on which RISC-V CPU you run it.
0
0
2
kudos to @ncopa again for the static busybox binary tip! i'll refine this at some point to multi-arch with toolchains from https://toolchains.bootlin.com/. i.e. one arg for arch and it will download the toolchain and matching busybox binary. #bootlin #buildroot
0
0
0
@kissane Maybe dude just to grow up a bit? :-) I'd factor this inverse to the age... I had a lot of silly ideas when I was barely adult (not necessarily on this topic tho but that does not matter) and I guess being somehow extreme is sort of part of growing up...
0
0
1
Jarkko Sakkinen
jarkko
Edited 1 year ago
Just wanted try out what I minimally need for qemu-system-x86_64 -kernel arch/x86/boot/bzImage:
#!/usr/bin/env bash
set -e
make defconfig
scripts/config --set-str CONFIG_INITRAMFS_SOURCE "initramfs.txt"
yes '' | make oldconfig
cat > initramfs.txt << EOF
dir /dev 755 0 0
nod /dev/console 644 0 0 c 5 1
nod /dev/loop0 644 0 0 b 7 0
dir /bin 755 1000 1000
slink /bin/sh busybox 777 0 0
file /bin/busybox initramfs/busybox 755 0 0
dir /proc 755 0 0
dir /sys 755 0 0
dir /mnt 755 0 0
file /init initramfs/init.sh 755 0 0
EOF
mkdir initramfs
curl -sSf https://dl-cdn.alpinelinux.org/alpine/edge/main/x86_64/busybox-static-1.36.1-r25.apk | tar zx --strip-components 1
cp busybox.static initramfs/busybox
cat > initramfs/init.sh << EOF
#!/bin/sh
mount -t proc none /proc
mount -t sysfs none /sys
sh
EOF
“linux from scratch” in 32 lines :-)
2
3
2
Jarkko Sakkinen
jarkko
Edited 1 year ago
Do not use Kate but since Krita has been so successful could be a good move 🤷 I mean Krita has been successful to the point that people know Krita but do not necessarily know KDE.
https://cullmann.io/posts/kate-on-all-platforms-2024/
https://cullmann.io/posts/kate-on-all-platforms-2024/
0
0
0
Jarkko Sakkinen
repeated
repeated
Jan Wildeboer 😷
jwildeboer@social.wildeboer.net
My B4 #OLPC prototype, 15+ years old. Original battery. Still works and battery charging is no problem. Impressive little machine :)
6
4
2
@kernellogger if you consider audio plugins they are actually just data processors so getting windows IO work right in Linux is the key. I don't see then actually any reason to do actual Linux ports of those plugins because with right stack underneath you could make the already existing Windows plugins perform better than in native Windows.
https://github.com/robbert-vdh/yabridge sort of solved the audio equation except for the IO part that requires better support from kernel. now we get that part right...
https://github.com/robbert-vdh/yabridge sort of solved the audio equation except for the IO part that requires better support from kernel. now we get that part right...
0
0
2
Jarkko Sakkinen
jarkko
@kernellogger hmm... is this inspired *BSD? have a faint memory that it had something similar but could remember wrong. nice addition anyway although personally ntsync driver is the thing for me in the next release :-)
2
0
0
Jarkko Sakkinen
jarkko
Edited 1 year ago
I've been strongly opinionated that CPU vendors are not paying the price of getting their confidential computing technologies to upstream but just made up analogue what goes wrong.
Think of desktop Linux. It is not business at scale but it is still mandatory piece for enabling business at scale.
Think of the world where Linux would run only on data centers but all development could be done only in Windows, and for testing, even when just playing with an idea, you would need a server rack with four figure price.
This is the world of confidential computing.
Think of desktop Linux. It is not business at scale but it is still mandatory piece for enabling business at scale.
Think of the world where Linux would run only on data centers but all development could be done only in Windows, and for testing, even when just playing with an idea, you would need a server rack with four figure price.
This is the world of confidential computing.
0
0
2
Jarkko Sakkinen
jarkko
@harrysintonen I wonder if they do something similar in Tampere University network. I get oauth2 tokens created with Thunderbird in Linux but not with GNOME Evolution or any more sensible other email client. So I don't use tuni address for email at all...
0
0
0
Jarkko Sakkinen
jarkko
With Linux 6.10 #ntsync driver combined with #wineasio you can get better #Windows than native Windows is for audio in some ways.
FL Studio is fast and smooth and you can do incremental backups for your full audio setup by simply archiving the wine folder. E.g. try a random plugin and revert back to previous wine folder after the experiment with all the crap cleaned up.
#audio #musicproduction #linux
FL Studio is fast and smooth and you can do incremental backups for your full audio setup by simply archiving the wine folder. E.g. try a random plugin and revert back to previous wine folder after the experiment with all the crap cleaned up.
#audio #musicproduction #linux
0
0
1
Jarkko Sakkinen
jarkko
Edited 1 year ago
When it comes "non-inclusive" language and even "toxic language", I try to keep somewhat high threshold before making any conclusions.
Sometimes a non-native speaker can do that by mistake because of less vocabulary and taking this into account is also one way of being "inclusive".
Usually when I get a response that feels just nasty the action I take is not to respond at all because when people get pissed trying to discuss about it just leads to non-constructive debate. Sometimes there's reason for that nasty response, e.g. a mistake that I've made or something I've ignored, but there are better and worse work weeks, and life outside the work affects the performance.
That is the reason why I think not responding at all to a "more toxic" response is the best action taken because also the person who responds might just have a bad week, and despite CoC and everything people sometimes cannot control their nerves - that's just human.
#linux #lkml #coc #codeofconduct
Sometimes a non-native speaker can do that by mistake because of less vocabulary and taking this into account is also one way of being "inclusive".
Usually when I get a response that feels just nasty the action I take is not to respond at all because when people get pissed trying to discuss about it just leads to non-constructive debate. Sometimes there's reason for that nasty response, e.g. a mistake that I've made or something I've ignored, but there are better and worse work weeks, and life outside the work affects the performance.
That is the reason why I think not responding at all to a "more toxic" response is the best action taken because also the person who responds might just have a bad week, and despite CoC and everything people sometimes cannot control their nerves - that's just human.
#linux #lkml #coc #codeofconduct
1
1
2
