klange changed the topic of #osdev to: Operating System Development || Don't ask to ask---just ask! || For 3+ LoC, use a pastebin (for example https://gist.github.com/) || Stats + Old logs: http://osdev-logs.qzx.com New Logs: https://libera.irclog.whitequark.org/osdev || Visit https://wiki.osdev.org and https://forum.osdev.org || Books: https://wiki.osdev.org/Books
thinkpol has quit [Remote host closed the connection]
thinkpol has joined #osdev
nyah has quit [Quit: leaving]
dza has quit [Remote host closed the connection]
tixlegeek has quit [Quit: tixlegeek]
vai has quit [Ping timeout: 240 seconds]
janemba has quit [Read error: Connection reset by peer]
zarock has joined #osdev
dude12312414 has joined #osdev
gog has quit [Ping timeout: 240 seconds]
dude12312414 has quit [Client Quit]
goliath has quit [Quit: SIGSEGV]
janemba has joined #osdev
netbsduser`` has quit [Ping timeout: 240 seconds]
heat has quit [Ping timeout: 252 seconds]
pretty_dumm_guy has quit [Quit: WeeChat 3.5]
masoudd has quit [Quit: Leaving]
dude12312414 has joined #osdev
bumbl3b33 has quit [Remote host closed the connection]
dza has joined #osdev
netbsduser`` has joined #osdev
skipwich has quit [Quit: DISCONNECT]
skipwich has joined #osdev
vdamewood has joined #osdev
netbsduser`` has quit [Ping timeout: 255 seconds]
* geist nops that dlclose
netbsduser`` has joined #osdev
vai has joined #osdev
<kazinsal> xd uwu rawr *nops ur dclose* :3
* kazinsal disappears in a puff of smoke
fkrauthan has quit [Quit: ZNC - https://znc.in]
fkrauthan has joined #osdev
netbsduser`` has quit [Ping timeout: 260 seconds]
[_] has quit [Remote host closed the connection]
netbsduser`` has joined #osdev
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
netbsduser`` has quit [Ping timeout: 258 seconds]
ThinkT510 has quit [Quit: WeeChat 4.0.5]
ThinkT510 has joined #osdev
<kazinsal> fiddling around with some design stuff because I really do need to get my idiot spare time project back on track and continue implementing my bespoke router OS
<kazinsal> and I've realized I've really got two options for the "user space" aspect, or at least, the syscall layer
<kazinsal> either A) do everything NIH and unique and all that, and thus not really feasibly be able to port any software and have to reimplement everyhing from scratch
<kazinsal> or B) shim in enough unix stuff to be able to have standard POSIX and libc stuff work natively so I can just port things like ospfd and bgpd and sshd
<kazinsal> so now my poor mangled brain is trying to figure out what the best approach is, if I should do more of a raw "implement unix syscalls" deal or if I should channel the spirit of the NT team and build some kind of intermediary subsystem layer where I can translate unixisms with hopefully minimal impact
<kazinsal> as much as the idea of doing a whole NIH implementation of every major protocol known to man is a fun one on paper, in practice, I really do not want to reimplement the insanely complicated state machines that run the whole internet
<kazinsal> but I also don't just want to write another unix
<kazinsal> unless it's for some stupid ancient system, because, y'know, peak dork hours
GeDaMo has joined #osdev
Left_Turn has joined #osdev
friedy has quit [Ping timeout: 240 seconds]
gabi-250 has quit [Remote host closed the connection]
gabi-250 has joined #osdev
gbowne1 has quit [Quit: Leaving]
gog has joined #osdev
vai has quit [Ping timeout: 248 seconds]
Burgundy has joined #osdev
tixlegeek has joined #osdev
kfv has joined #osdev
craigo has joined #osdev
tixlegeek has quit [Quit: tixlegeek]
yoyofreeman has quit [Remote host closed the connection]
yoyofreeman has joined #osdev
pretty_dumm_guy has joined #osdev
Burgundy has quit [Ping timeout: 255 seconds]
Left_Turn has quit [Ping timeout: 264 seconds]
TheCatCollective has quit [Quit: Meow Meow Meow Meow Meow Meow Meow Meow]
duderonomy has quit [Remote host closed the connection]
duderonomy has joined #osdev
Cindy has quit [Ping timeout: 255 seconds]
goliath has joined #osdev
zxrom has quit [Quit: Leaving]
kfv has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<mcrod`> hi
Left_Turn has joined #osdev
kfv has joined #osdev
kfv has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
yoyofreeman has quit [Remote host closed the connection]
yoyofreeman has joined #osdev
netbsduser`` has joined #osdev
duderonomy has quit [Remote host closed the connection]
kfv has joined #osdev
kfv has quit [Client Quit]
kfv has joined #osdev
kfv has quit [Client Quit]
heat has joined #osdev
[itchyjunk] has joined #osdev
Burgundy has joined #osdev
tixlegeek has joined #osdev
netbsduser`` has quit [Ping timeout: 246 seconds]
netbsduser`` has joined #osdev
sauce has quit [Ping timeout: 255 seconds]
sauce has joined #osdev
goliath has quit [Quit: SIGSEGV]
<mcrod`> interesting
<mcrod`> switching the toolchain build to 20.04 segfaults the LLVM build
<heat> "interesting"
Left_Turn has quit [Remote host closed the connection]
<mcrod`> yes, interesting
<zid> he's sworn off complaining, heat, give him some praise
Left_Turn has joined #osdev
<heat> noice
<heat> FWIW my toolchain builds run on ubuntu 20.04 and 22.04
<mcrod`> i'm waiting to see what happens
<mcrod`> i'm running it again
<mcrod`> might've just been a dumb fluke
<heat> where are you running the builds?
<mcrod`> buildjet ubuntu 22.04 runner through github
<mcrod`> er 20.04
<mcrod`> otherwise i'll wait 5 hours
<zid> buy github some ecc
<mcrod`> 2023-10-21T14:49:21.4271183Z /home/ubuntu/actions-runner/_work/SAMEthing/SAMEthing/toolchains/unix/staging/llvm-project-17.0.2.src/llvm/include/llvm/TableGen/Record.h:2006:8: warning: stack frame size (1099511627768) exceeds limit (4294967295) in 'llvm::RecordKeeper::addExtraGlobal(llvm::StringRef, llvm::Init*)' [-Wframe-larger-than]
<mcrod`> yep, it seems to completely shit out
<heat> what compiler are you using to compile?
<definitely-mjg> did you fail to optimize it?
<mcrod`> clang
<definitely-mjg> the compiler itself
<heat> is that LTO? reads like LTO
<mcrod`> yes LTO is enabled
<heat> disable it
<heat> problem solved thank you very much
<mcrod`> i'll be sad if that actually helps.
<definitely-mjg> noice
<definitely-mjg> o stands for overflow
<heat> also FWIW I can build clang LTO'd by ubuntu 20.04 clang just fine
<mcrod`> i dunno, maybe it's this runner
<mcrod`> this works perfectly fine on 22.04
<mcrod`> and I'm not doing anything crazy
<mcrod`> so
<heat> definitely-mjg, hi, you're our local ubuntu user, what do you think?
<definitely-mjg> i think LMAO
<mcrod`> i'm holding my cat for good luck
<definitely-mjg> sac a goat
<heat> in netbsd for the VAX, this wouldn't be a problem, because you'd run out of memory much earlier than this crash
<heat> follow me for more tips and tricks
<definitely-mjg> > Add a bunch of changes from the past 2 months that stood out to me.
<definitely-mjg> + vax: Reduce the object size of the malloc(3) code by converting
<definitely-mjg> + to rbtree, also improve performance and locking.
<definitely-mjg> priorities mon
<definitely-mjg> who gives a flying fuck their vm can't apply backpressure
<definitely-mjg> when one can optmize malloc on VAX
<mcrod`> heat: it's failing still
<mcrod`> do you use apt fetched clang or LLVM’s apt repo
<heat> apt
<mcrod`> fuckface
<mcrod`> okay
<bslsk05> ​lore.kernel.org: [linux-next:master] [drivers/char/mem] 1b057bd800: stress-ng.splice.ops_per_sec -99.8% regression
<definitely-mjg> (spoiler alert: not only no regression, but in fact a speed up)
<heat> yeah i saw that
<heat> it's a stress-ng bug, already fixed upstream
goliath has joined #osdev
zxrom has joined #osdev
Nixkernal has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
Nixkernal has joined #osdev
asarandi has quit [Remote host closed the connection]
Arthuria has joined #osdev
Arthuria has quit [Remote host closed the connection]
pretty_dumm_guy has quit [Ping timeout: 240 seconds]
Bonstra_ has quit [Quit: Pouf c'est tout !]
pretty_dumm_guy has joined #osdev
Bonstra has joined #osdev
xenos1984 has quit [Ping timeout: 260 seconds]
xenos1984 has joined #osdev
dude12312414 has joined #osdev
Arthuria has joined #osdev
kfv has joined #osdev
Arthuria has quit [Remote host closed the connection]
kfv has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Bonstra has quit [Quit: Pouf c'est tout !]
Bonstra has joined #osdev
Matt|home has joined #osdev
<geist> oh neat someone is trying to fix vax issues
<geist> also oh nice, splice on /dev/zero would be lovely
<heat> what's the usecase?
<definitely-mjg> if this was not #osdev i would assume geist is being sarcastic
<geist> oh i use this all the time for benchmarking and overwriting disks and whatnot
<geist> i love the `pv` utility, which acts like cat but also prints the progress
<geist> but it *also* uses splice
<heat> oh
<geist> so for example `pv /dev/zero > something` would actually achieve a no copy path
<geist> i have straced it before and noticed the /dev/zero wouldn't splice, so it falls back to doing a read/write
<geist> does it matter really? not really, but doing it with zero copy pleases me
<definitely-mjg> i think this may be the case of missing the point
<definitely-mjg> like vdso-based getpid
<definitely-mjg> for benchmarking
<geist> well, in the case of something like pv /dev/zero | md5sum or whatnot you'd like to factor out the overhead of pv or dd or whatnot
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
craigo has quit [Quit: Leaving]
xenos1984 has quit [Ping timeout: 260 seconds]
<definitely-mjg> hm
<definitely-mjg> well i defo did not think of that kind of use cae
<heat> good newz: my page reclamation code works-ish-kinda, at least leaves the system in a surviving state
<definitely-mjg> against what
<definitely-mjg> that find thing?
<heat> no
<definitely-mjg> cause you may be outdoing netbsd soon mate
<heat> something a lot simpler
<heat> page_fault2 -t4 on 1G memory
<bslsk05> ​mail-index.netbsd.org: Re: Unexpected out of memory kills when running parallel find instances over millions of files
<definitely-mjg> and a followup
<definitely-mjg> there's been new replies
<definitely-mjg> that's a lame test mon
<geist> come on, again
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
<geist> there are folks here that might actually work on netbsd, etc
Matt|home has quit [Quit: Leaving]
<heat> i know it's a lame test
<heat> i have a lame system
<definitely-mjg> would you even survive creating the filez
<heat> no
<definitely-mjg> [x] lame [ ] not lame
<definitely-mjg> indeed
Left_Turn has quit [Remote host closed the connection]
<heat> actually disregard that, it falls over pretty easily
<definitely-mjg> do you need an "extra lame" form?
Left_Turn has joined #osdev
<heat> sorry I only take feedback from people who actually wrote their own OS
<geist> hmm, what was that inexcusably lame meme
<definitely-mjg> heat: i did, kind of
<definitely-mjg> heat: got a kernel with printf and some exception handlers
<definitely-mjg> does it count
<heat> no
<heat> that sounds *mega lame*
xenos1984 has joined #osdev
<bslsk05> ​'wayne's world suck cut.MPG' by Jared G (00:00:26)
<geist> there was a waynes world skit somewhere where i think they yell at the camera 'inexcusably lame!' but i cant find it
<heat> how slow are indirect calls?
<geist> like, in general?
<definitely-mjg> bro
<definitely-mjg> SPECT-FUCKING-RE
<heat> in general, but also retpoline
<definitely-mjg> they are unbearably sucky
<definitely-mjg> tolerable otherwise
<definitely-mjg> enough to not work around them
<heat> what
<geist> an in general anything modern can cache the shit and prefetch through indirects, especially if they mostly take a specific path
<geist> repoline of course destroys that, which is the point
<heat> gotcha
<geist> also of course hard to benchmark because the harder you wail on it the faster it gets
<heat> i bet this gets lost in the noise anyway
<geist> due to cache
<definitely-mjg> inb4 heat decides to indirect locking primitives like this
<geist> i tend to try to avoid them in critical paths if possible, but for generic stuff, meh
<geist> right, like indirecting locking primitives is probably not a great idea
<definitely-mjg> instaban
* geist is playing weith metadata caching on their nas box
<heat> nah, i'm talking about struct vm_ops kind of stuff
<geist> yeah i think that's a case where it's just necessary
<definitely-mjg> what vm ops, you mean for backing store?
<geist> any other solution is probably worse
<geist> since your vm objects are almost certainly polymorphic by design
<definitely-mjg> so for example nobody is going to complain that you have an ops table for struct file
<definitely-mjg> or inodes
<heat> i used to have a spaghteti-ish code path that handled anonymous/private/shared
GeDaMo has quit [Quit: That's it, you people have stood in my way long enough! I'm going to clown college!]
<geist> yeah and keep in mind a bunch of if/elses is also relying on the branch predictor
<heat> i'm moving this to a struct vm_ops { int (*fault)(struct vm_pf_context *); };
<geist> though branch predicting a simple cmp/branch isa bit easier
gbowne1 has joined #osdev
gbowne1 has quit [Remote host closed the connection]
gbowne1 has joined #osdev
Left_Turn has quit [Ping timeout: 258 seconds]
<heat> [pedro/vm 5ab2b2d7] vm: Remove old MAP_PRIVATE paths
<heat> 4 files changed, 4 insertions(+), 350 deletions(-)
<heat> this is what you get when you remove vm_object COW-ing
<definitely-mjg> write a freebsd variant
<heat> no
* definitely-mjg goes bakck to shitting on onyx
<heat> i would rather write a serenityOS clone
<heat> and i hate windows 95
Left_Turn has joined #osdev
kfv has joined #osdev
<definitely-mjg> win 95 more stable than onyx
<heat> probably yeah
kfv has quit [Quit: Textual IRC Client: www.textualapp.com]
<definitely-mjg> yo geist
<definitely-mjg> check out the e-mail i got, no joke
<geist> sup
<definitely-mjg> re that netbsd OOMing stuff
<definitely-mjg> I think I reported this something like 20 years ago, but noone really seemed to care. I noticed it pretty much right away after NetBSD switched to the unified memory thing, where all free memory usually was grabbed as disk cache. It was not fun on VAX, but at the time it seem other platforms didn't suffer enough to consider it a problem. I guess over time it's just gotten worse...
<definitely-mjg> :p
<definitely-mjg> VAX
<heat> if only had they listen to the VAX users
<definitely-mjg> ikr
<geist> cool, nice that someone is looking at it. i was thinking of fiddling with trying to fix some tip of tree vax issues too
<geist> i dont think it's been stable on vax for a few major revisions
<definitely-mjg> this is not a vax problem, dude just mentions it
<geist> hmm?
<bslsk05> ​mail-index.netbsd.org: Unexpected out of memory kills when running parallel find instances over millions of files
<definitely-mjg> dude says this was the case since switch to uvm
<heat> mjg has a funny test that looks up millions of inodes and netbsd fails the test by OOM killing vs flushing old inodes out
<geist> ah
<geist> yah this is one of the interesting reasons why old platforms can be kinda useful, since they stress the system out in interesting ways
<definitely-mjg> it's not a funny test, it's an actual workload
<geist> so someone privately sent you an email about vax?
<geist> oh that they had seen the same problem on vax. now i get it
<definitely-mjg> VAX KURWA
<geist> honestly i dunno why you're so gung ho about trying to troll me on this
<definitely-mjg> i find in the same vein as people chanting RUST or ITANIUM
<definitely-mjg> happy to gtfo on that one
<geist> i'm seriously not going to bite, because i think it's geniunely nice that some folks are still fiddling with it. that's basically the *point* of netbsd, run on old things
<geist> if you want a more modern thing, run something else
<definitely-mjg> oh i totally agree
<definitely-mjg> wanna fuck around with old archs, netbsd is there for you
<geist> yep, and conversely i am 100% behind linux dumping arches like itanium or even x86-32 when the time comes
<definitely-mjg> i am taking stabs at it because it pretends to be viable on actual hardware contemporary hardware
<geist> old arches do hodl you back
<heat> geist: :(
<definitely-mjg> ye agreed
<geist> well okay, i'm like 60% behind. at least the conversation should be had
<definitely-mjg> quite frakly i would love if anything non-64 got whacked
<heat> ya know ARM is still selling 32-bit right? and so is riscv
<geist> somewhat agree, though 32bit will be strnog and alive for like 20 more years, since it's perfectly viable in emdedded world
<geist> but x86-32 yes
* mcrod` thumbs up
<heat> in fact, ARM sells ya boards with > 4GB ram
<definitely-mjg> i am aware :(
<definitely-mjg> just sayin, 's a constant fucker
<definitely-mjg> being kva starved
<geist> but at least on those arches, riscv especially, the 32bitness is more of a detail, not a major rethink of things
<geist> but yeah it does impact the use of kva
<geist> which can and does have major design implications
<heat> kmap is cursed
<definitely-mjg> just implement 0 cost kmap in rust
pretty_dumm_guy has quit [Quit: WeeChat 3.5]
<heat> exception-based templated metaprogramming zero-cost kmap
<definitely-mjg> ok mjg@
<geist> ruuuuuust
<mcrod`> *FINALLY*
<geist> okay i just rebuilt my ssd cache on my nas box, this time pinning all of the btrfs metadata in it
<mcrod`> FUCK
<heat> but srsly if you told me like "SLAB and page cache can only be in lowmem" that's like, totally fine
<heat> it's still ~800MiB of memory
<geist> heat: not on VAAAAAX!
<heat> heck, my shitnix can only access 800MiB of memory
<geist> yeah that's some hot shit
<mcrod`> I'm going to try and port my OS when the time comes to... my packard bell
<mcrod`> but I think that'd be a huge waste
<definitely-mjg> you have an os?
<heat> no
<geist> mcrod`: oh that's the fun stuff. i literally bought a duplicate of my old 386 i had when i was a kid off ebay a few years ago
<definitely-mjg> heh, nice
<geist> so i could port stuff to it and then feel the cycle is complete
<mcrod`> i did the same thing
<mcrod`> minus "when I was a kid"
<definitely-mjg> actually i got you beat, kind of. my 486 is still in my parent's basement
<definitely-mjg> but probably does not work ;)
<mcrod`> yeah my packard bell is a 486
<heat> yeah i still have my old computer, might not work tho
<geist> yeah i was totally 'lucky' that one day i was looking at 386 and the exact model was just sitting there + a 5.25 drive
<heat> one of the 32-bit pentium 4s
<geist> so i yoinked it for like $50
<definitely-mjg> noice
<definitely-mjg> how ram you got there
<definitely-mjg> how much
<geist> me? 4MB!
<definitely-mjg> was gonna write if more than 4 then it's not period accurate
<geist> actually have an early slakware linux on it, it.... runs
<definitely-mjg> high end shite mate
<geist> linux 1.2 uses like 2.5 of the 4MB, so it's swapsville by default
<heat> ok geist now get the last 386-supporting linux kernel and use it
<definitely-mjg> i was sporting 8 on my 386
<definitely-mjg> erm, 486
<heat> probably lands you in linux 3.x? mabe 4.x
<geist> for shits n giggles i put a floppy in the 5.25 drive and set it as a swap device while installing from 3.5 disks
<mcrod`> I'm trying to find a video of my packard bell screaming
<geist> netted me another 1.2
<geist> but i also have a 486 with 32MB so that is good for running slightly later stuff. a nice am486-133 too, so basically as good as they got to, short of overclocking to 150 or 166
<heat> gosh why are you people sooooooo oolllllllllld
<mcrod`> I caution you, seriously; volume warning
<geist> heat: did you just realize you're sitting in the same room with a bunch of olds?
<heat> yes, all the oldies
<heat> all outdated
<mcrod`> to make matters worse
<mcrod`> after that terrible sound, it beeps 3 times, then 2, then 1
<definitely-mjg> 486 with 32M is some serious shit
<geist> yeah i kinda explicitly went in search on ebay of the highest end 486, just to have it
<definitely-mjg> which model is it
<geist> oh it's some random ass thing, from bulgaria, literally
<definitely-mjg> oh 133
<definitely-mjg> ok
<geist> yah am486-133
<definitely-mjg> i think that is outdone a pentium90
<definitely-mjg> by
<heat> 486 has invlpg therefore OPTIMAL
<geist> yep. in college in 1995 my roommate and i jointly assembled an am486 with 8MB and ran it as a headless linux box for a few years
<geist> it was a nice cheap but contemporarily fast machine at the time
<definitely-mjg> mon
<definitely-mjg> i had a 486 dx2 50 mhz
<geist> maybe 16MB. good enough for decent command line usage
<definitely-mjg> a friend of mine had 66 mhz
<definitely-mjg> and his box was running ricles around mine
<definitely-mjg> i had no idea why at the time, now i suspect he had caches
<definitely-mjg> circles
<geist> at the time just having a dedicated linux machine that you didn't have to dual boot to play your games and wasn't frustratingly slow was a huge win
<definitely-mjg> :D
<geist> like you were a hella turbo nerd at that point just *having* a linux box
<definitely-mjg> rebooting to windows to google for error messages == $$
<heat> linux in 95?
<geist> heat: yeah woulda been slackware for sure. probably around linux 1.0 timeframe. i remember being aware at least when 1.0 was announced
<geist> from whatever the .99 it was at before
<mcrod`> my first experience ever
<mcrod`> was ubuntu warty warthog
<geist> i think 1.2 was a few years later, and when it switched to ELF or whatnot. also the libc5 vs glibc wars were being fought i think
<mcrod`> and that was in 2004
<geist> now i work with roland every day so i can get the straight dope
<mcrod`> i actually had to order a CD
<geist> yah cdroms were still kinda rare in 95, thoguh i had some old mitsumi drive. somethingl ike CD33 or whatnot
<heat> wow that's really old
<geist> needed a custom card and a linux driver for it
<geist> to install something like slackware i just had to take a few boxes of disks to the computer lab on campus and punch them out one at a time, and when i got back to the dorm hope there weren't any read errors
<geist> about a 50/50 chance you'd make it through a box of disks without a corruption
<geist> late year 3.5 disks/drives were pretty shitty, honestly
<definitely-mjg> 3.5 disks were great
<definitely-mjg> 5.25 was the dogshit
<definitely-mjg> you had to handle it like a fucking mona isa
<geist> eeeh, i think the quality fell off over time, and the higher the density the more chance of failure
<definitely-mjg> well it may be my faulty memory, but i don't recall any 3.5 fucking up on me
<geist> i found over the years that old 720k disks are almost entirely fine, but 1.44s fail pretty quickly
<geist> but that's a density issue
<definitely-mjg> and i messed with them a lot on atari and amiga
<geist> probably lower density
<definitely-mjg> probably
<mcrod`> heat: on the packard bell I have, the CD-ROM drive's data cable is connected directly to the sound card, which plugs into a daughterboard, which plugs into the mobo
<geist> mcrod`: yeah that was super common, lots of sound cards had a cdrom cable
<mcrod`> still crazy to me
<geist> i forget precisely if those were just IDE adaptors, in the end adaotpr slot (since lots of smallish machines at the time only had one 40 pin connector on the mobo) or something else
<geist> since an IDE 'controller' is not much but an ISA bus pass through it wouldn't be hard for the sound card to just go ahead and add another one
<geist> with a jumper or whatnot on the card to disable in case your mobo actually had a second 40 pin IDE adaptor
<definitely-mjg> i think that was for audio tracks?
* definitely-mjg had one as well
<geist> for audio there wasa little 3 pin cable from the cdrom to the sound card
<definitely-mjg> genz don't know that games would fit on a cd so comfortably that game soundtrack would be literal tracks
<definitely-mjg> ye
<geist> the 40 pin connector on the sound card was mostly so they could sell a sound card + cdrom combo thing, since lots of machines at the time only had a single 40 pin connector for hard drive
<geist> and ou dont really want to share a HD with a cdrom, at least at the time
<geist> the cdroms are so slow and occupy the bus it really hurts hd performance
<sham1> Hello, World!
k0valski1889162 has quit [Quit: Ping timeout (120 seconds)]
<geist> hola
<mcrod`> christ
<mcrod`> well, never mind.
<geist> hmm, wasn't there an old meme 'christ, what an asshole'
<definitely-mjg> i don't know if i'm not old enough
<geist> like you could take any comic and replace the text with that
<definitely-mjg> or not very meme-y
<bslsk05> ​knowyourmeme.com: Christ, What An Asshole | Know Your Meme
<geist> well mid 2000s
<definitely-mjg> let's say older than heat is moderatly old
<geist> yeah that's like mid-old
<definitely-mjg> you reminded me of train etiquette
<definitely-mjg> that was top notch nsfw memes
heat has quit [Remote host closed the connection]
heat has joined #osdev
sbalmos has quit [Remote host closed the connection]
sbalmos has joined #osdev
<heat> i just saw the greatest thing when it comes to vm region splitting that i now can't unsee
<heat> my logic was 150 lines of HAIRY BROKEN MESS
<heat> it turns out that if you're smart and tidy you can have a helper function + at most 2 splits and that handles all the cases
<geist> yeah that's a gnarly routine to write optimally
<Ermine> definitely-mjg: genz don't know that games would fit on a cd so comfortably --- That's why doom soundtrack was some weird MIDI variation?
<Ermine> And that's why RTW soundtrack was combined into a single file?
<geist> yah some of them had sountracks on the CD. like racing games
<geist> also was a trick PS1 did a bunch too
<Ermine> Rise Of Nations had its soundtracks as actual tracks, even with proper names
<Ermine> Speaking of RTW, it has a lot of stuff configurable with text files, and it feels nice
<Ermine> I guess true nerds of that time were playing as SPQR or Rebels or Armenia
<zid> SPQR!
<zid> I had that as my discord icon for aaages
<mcrod`> i never thought i'd see excitement from you in my life
<zid> well none of you talk about honzuki
<zid> just boring BSD stuff
<zid> next vol is going to be fucking rad btw
<definitely-mjg> `what is even honzuki
<definitely-mjg> high school anime?
<clever> definitely-mjg: https://myanimelist.net/anime/39468/
<bslsk05> ​myanimelist.net: Honzuki no Gekokujou: Shisho ni Naru Tame ni wa Shudan wo Erandeiraremasen - MyAnimeList.net
<definitely-mjg> am i going to land on a list if i open that
<clever> nope
<nortti> you will end on myanimelist at least, but I don't think any others
<definitely-mjg> that's some game of thrones stakes over there
sbalmos has quit [Quit: WeeChat 4.0.5]
<zid> why would you link to mal
<zid> when wikipedia etc exist
Left_Turn has quit [Read error: Connection reset by peer]