pabs3 has quit [Read error: Connection reset by peer]
pabs3 has joined #osdev
agent314 has joined #osdev
skipwich has quit [Remote host closed the connection]
skipwich has joined #osdev
elderK has quit [Quit: WeeChat 4.6.3]
agent314 has quit [Read error: Connection reset by peer]
agent314 has joined #osdev
agent314 has quit [Max SendQ exceeded]
agent314 has joined #osdev
agent314 has quit [Max SendQ exceeded]
agent314 has joined #osdev
agent314 has quit [Max SendQ exceeded]
stolen has joined #osdev
griddle has joined #osdev
agent314 has joined #osdev
_whitelogger has joined #osdev
eluks has quit [Remote host closed the connection]
eluks has joined #osdev
chiselfuse has quit [Remote host closed the connection]
gildasio has quit [Remote host closed the connection]
chiselfuse has joined #osdev
gildasio has joined #osdev
jcea has quit [Ping timeout: 248 seconds]
griddle has quit [Quit: griddle]
goliath has joined #osdev
rez has joined #osdev
_whitelogger has joined #osdev
netbsduser` has quit [Ping timeout: 258 seconds]
netbsduser` has joined #osdev
valerius_ has joined #osdev
skipwich has joined #osdev
j00ru has joined #osdev
netbsduser`` has quit [Ping timeout: 258 seconds]
netbsduser`` has joined #osdev
netbsduser` has quit [Ping timeout: 258 seconds]
rorx has joined #osdev
melnary has joined #osdev
goliath has joined #osdev
jeaye has joined #osdev
GeDaMo has joined #osdev
chromoblob has quit [Ping timeout: 248 seconds]
chromoblob has joined #osdev
chromoblob has quit [Ping timeout: 258 seconds]
ring0_starr has quit [Ping timeout: 250 seconds]
skipwich has quit [Ping timeout: 265 seconds]
netbsduser`` has quit [Ping timeout: 256 seconds]
Lucretia has joined #osdev
<zid>
Poo.
<Ermine>
poonix
Arsen has joined #osdev
qookie has joined #osdev
agent314 has quit [Ping timeout: 260 seconds]
chromoblob has joined #osdev
stolen has quit [Quit: Connection closed for inactivity]
agent314 has joined #osdev
chromoblob has quit [Read error: Connection reset by peer]
chromoblob has joined #osdev
alice has quit [Quit: *disappears* U_U]
gog has joined #osdev
gog has quit [Quit: byee]
alice has joined #osdev
agent314 has quit [Ping timeout: 248 seconds]
pepelegal has joined #osdev
skipwich has joined #osdev
edr has joined #osdev
Terlisimo1 has quit [Quit: Connection reset by beer]
Terlisimo has joined #osdev
jcea has joined #osdev
HeTo has quit [Remote host closed the connection]
puck has quit [Ping timeout: 250 seconds]
puck has joined #osdev
<ddevault>
man I am carrying a lot of boilerplate around for osdev projects
<Ermine>
time for libOS
<ddevault>
nearly 13,000 lines of code to vendor in the bootloaders I've been carrying around, configure the CPU the way I like, and get to an infinite loop in kmain
<kof673>
you could maybe say reverse, a charset is a collection of constant name/value pairs, but enough philosophy :D
jcea has quit [Ping timeout: 248 seconds]
vdamewood has joined #osdev
<MelMalik>
xD
chromoblob has quit [Read error: Connection reset by peer]
chromoblob has joined #osdev
goliath has quit [Quit: SIGSEGV]
gildasio has quit [Ping timeout: 272 seconds]
gildasio has joined #osdev
libercv has joined #osdev
svm has joined #osdev
msv has quit [Ping timeout: 248 seconds]
les has quit [Quit: Adios]
les has joined #osdev
libercv has quit [Quit: Konversation terminated!]
les has quit [Quit: Adios]
les has joined #osdev
skipwich has quit [Ping timeout: 248 seconds]
skipwich has joined #osdev
da5id has joined #osdev
netbsduser`` has joined #osdev
guideX_ has joined #osdev
guideX has quit [Ping timeout: 260 seconds]
guideX_ has quit [Ping timeout: 248 seconds]
guideX has joined #osdev
guideX_ has joined #osdev
guideX_ has quit [Remote host closed the connection]
guideX has quit [Ping timeout: 244 seconds]
vdamewood has quit [Quit: Life beckons]
guideX has joined #osdev
pogchamp has joined #osdev
pogchamp is now known as gog
agent314 has joined #osdev
vdamewood has joined #osdev
vdamewood has quit [Quit: Life beckons]
GeDaMo has quit [Quit: 0wt 0f v0w3ls.]
<sortie>
Today in osdev: I'm installing GNU Hurd. That feels like an upgrade.
<sortie>
(from doing html stuff)
<sortie>
... why is the GNU Hurd install saying NetBSD 10.99.12 (RUMP-ROAST)
<kof673>
it uses netbsd rump i guess :D
<sortie>
Apparently that's what a rump kernel is :D
<sortie>
Just my mild shock when Hurd claimed to be netbsd
<sortie>
Now we're playing my favorite game of 'Is Hurd solving the halting problem or is it just extremely slow'
<kof673>
i am guessing you want to test posix compliance/etc. on hurd
<sortie>
https://sortix.org/os-test/ ← Yep. I just published my new os-test include suite that verifies whether the system headers has the declarations required by POSIX.1-2024. I'm updating each of the operating systems right now. Debian Hurd seems to be so poorly supported that I might have to reinstall
<bslsk05>
sortix.org: os-test
<kof673>
that is a guess, not prying, just why else install hurd j/k
<geist>
oh yeah i was meaning to do the same thing with the new hurd release
<ThinkT510>
sortie: hurd does have x86_64 support now. undoubtedly you'd need a fresh install for that
<sortie>
ThinkT510: I did try a x86_64 build last time and it didn't work. I wanna see if I can upgrade my i386 install
<sortie>
Alright I hacked my hurd debian apt sources and gpg keys to point to the new release.. let's see if apt-get dist-upgrade works or how much this explodes in a fun manner (I backed up my disk image)
<zid>
hurd has support for anything at all? :o
<sortie>
Hurd does work, more or less, although the debian support is a bit of an afterthought
<ThinkT510>
ha, it is like their main distro
<ThinkT510>
unless you'd like to get their guix version going
xenos1984 has quit [Read error: Connection reset by peer]
<sortie>
I mean, more like, I got a GNU hurd install from last year and the apt stuff has bitrotted and doesn't work anymore. I'm not seeing a supported path to do a network upgrade of my Hurd server from the old version to the latest version. That's a deal breaker in production stability
<sortie>
Meanwhile my reinstall attempt has been locked up for half an hour, so rebooting now
<ThinkT510>
do you try to give similar specs to the VM for each OS you test or do you try to stick to only what upstream recommends?
<sortie>
I just put them in a standard default qemu with some light-to-moderate resources unless they needed anything special
<sortie>
It's all scripted up and integrated with systemd on my server. It brings up the virtual machines using my scripts, and ssh's in and powers off the virtual machines when systemd stops the unit
emneo has joined #osdev
strategictravele has joined #osdev
Turn_Left has joined #osdev
Left_Turn has quit [Ping timeout: 260 seconds]
xenos1984 has joined #osdev
strategictravele has quit [Quit: strategictravele]
strategictravele has joined #osdev
<sortie>
Weird. There was some kind of crash during my hurd upgrade. I think hurd itself restarted or something. The system is still kinda alive but really, really, really slow, like CPU or IO starvation.
<heat>
hurd? crashing?
<sortie>
I wonder if that's the problem I'm seeing when trying out the new hurd where it seems to go extremely slow after configuring the keyboard
<heat>
but it's a microkernel
<heat>
i heard those won't crash
<sortie>
Honestly so far the microkernels I've tried are Minix and Hurd.. and I think Tanenbaum was just making up claims lol
<sortie>
Well my bastard hybrid upgraded hurd has become so slow I have to forcefully reboot it. At least the grub upgrade went through but who the fuck knows if this frankenstein system boots.. let's find out :D
<heat>
i want to bring this to your attention my dear european citizen
<heat>
THEY ARE GETTING EUROPEAN FUNDING
<heat>
BUCKETS OF IT
<sortie>
and well now it's stuck after doing some ACPI stuff
<sortie>
This disk image is probably hosed
<sortie>
Sooo 1) installing the new Hurd i386 locks up after doing the keyboard configuration 2) upgrading my old Hurd i386 install crashes midway and hoses the system
<sortie>
Time for 3) let's try a fresh x86_64 install. Maybe it actually works this time.
ring0_starr has joined #osdev
ring0_starr has quit [Max SendQ exceeded]
ring0_starr has joined #osdev
<nikolar>
i am pretty sure tanenbaum didn't invent the concept of a microkernel
<nikolar>
or what their advantages are supposed to be
<heat>
andrew microkernel vs linux torvalds
<sortie>
heat: To be honest, a lot of the experience building os-test and installing every single OS and upgrading them -- it very much inspired me to do better and made me go 'oh I can definitely do a better experience'
<sortie>
It's part of why I'm doing the funding game now
<gog>
but is trianglix going to be the future
<gog>
feel the power of the triangle
<sortie>
gog: Trianglix is already the past, present, and future. All three edges!
<gog>
wow 3edgy5me
<heat>
i think we found john userexperience
<heat>
sorry jonas userexperience*
<sortie>
I welcome the archetype
<sortie>
It might actually be a better one than Stan Dard that I wrote
<sortie>
I.. say that but literally spent weeks enforcing standards on all operating systems
<sortie>
heyyy Hurd x86_64 actually gets past asking for a keyboard layout
<gog>
ooh ooh what's my osdev stereotype
<gog>
i think i'm a vera vaporware
<sortie>
So far Hurd is at 78% detecting filesystems for a minute inside the debian installer. There's literally only their .iso and an empty harddisk.
<sortie>
And this is literally just a default qemu invocation
<Ermine>
sortie: you can try to kick hurd with nmis to keep it alive
<Ermine>
that worked with minix
<sortie>
god
<sortie>
plz no
<sortie>
that was evil
<sortie>
oh maybe it's running out of memory. I did only 1 GB it
<Ermine>
:D
<sortie>
The debian guide did say 2 GB
<nikolar>
what's nmis
<Ermine>
non-maskable interrupts
<nikolar>
ah
<nikolar>
NMIs
<Ermine>
you can fire it with qemu's nmi command
<Ermine>
not in mood to press shift key
<sortie>
Minix had bitrotted to the point where it didn't actually work in qemu out of the box. I had to poke it with nmi a couple times during the install. At least it worked afterwards
<nikolar>
kek
<sortie>
Alright it made it to partitioning now
<Ermine>
hurd has a dude working on it at least
<sortie>
Honestly my big problem with GNU in general is that nobody is in charge of quality
<sortie>
Like, it's not some person building a nice .iso and making sure it works for everyone
<sortie>
But rather just the Hurd people doing hurd, and somehow guix and debian packages that and maybe makes it work who knows
<sortie>
Just so many layers of indirection between the devs and people apparently supposed to run this thing
<sortie>
It's 2025 and this is what they've built so far after all these years? But hey, it does work kinda more or less. I think.
<nikolar>
you do realise it's basically a hobby project right
<nikolar>
lol
<Ermine>
add words "amateur OS" and everything becomes clear
<sortie>
I also do officially dislike how the debian graphical installer doesn't say what's it going between prompts. Just a blue screen and a blicking cursor.. like I'm not sure what's going on
<sortie>
nikolar: Totally :P BUT.. I mean. GNU is supposedly a serious organization that does a big and professional OS :P
<nikolar>
say that again when anyone at all considers hurd a serious os
<sortie>
I am happy that Hurd is alive though :) It's a shame that Minix appears to have died
<Ermine>
that was true in 90s?
<sortie>
uh so
<sortie>
So far my Sortix installation experience is vastly superior to Hurd and also works
<sortie>
Sending a nmi makes the hurd output '1' to the console lol
<nikolar>
1
strategictravele has quit [Quit: strategictravele]
<sortie>
And so far Hurd the installer is stuck on a blue screen with a black flashing cusor for several minutes with no explanation
<sortie>
And that's trying to do a fresh install of their new x86_64 port. The i386 port didn't even make it past the keyboard layout config
<sortie>
In a standard modern qemu with no special options and plenty of resources
<Ermine>
i care most for the desktop, and neither sortix nor hurd experience are satisfactory to me
<sortie>
It's honestly a shame that Hurd seems less stable than it was in 2023..
<nikolar>
oooh
<nikolar>
fight fight fight
<sortie>
Ermine: I hear ya. I think Haiku is the best contended for the user experience in the osdev world!
<sortie>
I was very very impressed with how they had more-or-less smooth Youtube playback via the wifi on a real laptop at FOSDEM
<sortie>
Ermine, you should note that my intended user experience for Sortix is a server user experience, not a desktop user experience. I think I can do very, very well there with modest resources. The client desktop experience is much much harder a game, especially in terms of drivers, and I'm intentionally not competing there at this time
emneo has quit [Quit: WeeChat 4.7.1]
<sortie>
I was also very pleasantly impressed with how much the OpenBSD upgrade experience has improved. I just had to run 'sysupgrade' with no arguments and it took care of it for me, even rebooted. I did have to manually run 'sysmerge' afterwards (although I had no conflicts, so wasn't needed) and pkg_add -u (to update my packages). The most annoying part is that I still have to delete old files in the base system manually.
<sortie>
The Sortix upgrade experience was very much built out of frustration with how much manual action was needed to upgrade OpenBSD. My 'tix upgrade' experience is still better, but OpenBSD is starting to have the best upgrade experience of the BSDs no question.
<sortie>
FreeBSD was quite smooth too, a bit more complicated in other ways, it did ask about details I didn't care about, but otherwise a solid grade too
<sortie>
NetBSD was just.. bad. I had to manually construct an URL and give it to it. The sysupgrade tool wasn't even shipped with the system and I had to get it from pkgsrc. Then I had to invoke a bunch of commands. It then started updating my /etc and asked whether it could replace /etc/passwd.master and delete my root and os-test users. Then it asked me about a bunch of configuration files I didn't care about. And a lot more. Overall, a poor grade. But it worked.
<sortie>
DragonFlyBSD's upgrade experience, apparently, is asking you to reinstall or build from source. Failing grade.
<Ermine>
> server experience
<Ermine>
Ok so I want to build android. It's good to build such heavy stuff on servers
<Ermine>
so i need *check notes* 64G of ram and lots of cores
<sortie>
and so far I've spent two hours trying to upgrade Hurd to no success. A big shame, I was really looking forward to seeing the latest version in action
Left_Turn has joined #osdev
<sortie>
Haiku also just improved their POSIX conformance score to 90% after I upgrade to the latest nightly build. Impressive. They're actually ahead of OpenBSD and macOS