Notices by nina (q66@blahaj.social)

@lanodan it literally stayed in rolling boil for like 2 minutes as i took it off

smelled very bad too

>left moka pot boiling on the stove for the past half an hour

extremely good job me

@dalias @whitequark it reads as "you fools can do whatever you want as long as the thing you're protesting against is unaffected" which is total clown shit

@mia @0x9E01 the whole idea behind overcommit is scuffed and it's unlike what basically every other os does and it makes it impossible to trust the allocator to reliably fail

you can get less sucky *behavior* out of it with careful tweaking (and ideally a cgroups-aware userspace oom daemon) but it won't make it any less busted

@domi https://gist.github.com/q66/469a7f778624a35faa01ed6243e890dd this is the contents of the vendor dir for 1.84

(the checksum clearing is because if you patch a vendor crate cargo will yell at you because it keeps checksums of all the files, so the solution is to just clear the json file and then it stops complaining)

@domi it's literally every libc crate version rustc vendors and builds

@april @domi puters provide me with this little hyperspace where i can pretend i'm not completely useless + maybe on a good day even make some entity's life a tiny bit better so that is nice i guess

@koakuma so did they just drop it with no replacement? do they have a journal

@koakuma big oof

openbsd moment

aspeed vga does not like big endian too much lol

hmmm should i test the intel gpu on this

yea figures, firmware load fail

maybe if i'm really bored sometime i might test how far i can take this

@linear it works great on little endian already (got an a750 and a380 and they are both stable, planning to buy a b580 as soon as they are actually available here and maybe then once i have a spare i can take a look)

@linear well that's basically nothing

the only stuff that works semi-reliably at all is old amd terascale (anything older is much bigger jank, anything newer does not work at all) and nowadays you also need mesa-amber because mesa is bust as well

@linear well now with uh, certain people newly doing major haphazard refractors in mesa on regular basis you can expect things to remain in a state of perpetual brokenness pretty much

given how the bottom side of the board is a forest of tiny smd thingys i'm mostly just glad i did not accidentally scrape off some during the time it was just laying around (or if i did do that it was not important enough to prevent boot)

i will probably just set this up as a build machine later so the poor osuosl vm does not have to build two architectures at once

things to do at 5am: get the power9 blackbird board i had not booted in 1.5 years and which had a weird busted up custom bmc plus custom-built host firmware working again

put into a spare case i had around, get bmc serial console, magic trick to disable u-boot watchdog, shave some yaks (package a tftp server for chimera), netboot the image-bmc with temporary rwfs, rebootstrap everything with stock images, and we are so back

>update ca-certificates
>try to push into cports
>remote rejected
>"wow, did something in the certs break"
>nope, it's just github going down that exact minute

phew

wtf came back home and there was silence and darkness

apparently the electricity was off, had to flip the breaker back on

no idea how long, freezer was totally melted so everything in there is bust

butter in the fridge still solid tho and my sourdough starter is intact too...

wtf the corunya train station is also entirely remodeled

they were busy in spain while i was gone