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
gioyik has joined #osdev
[itchyjunk] has quit [Ping timeout: 258 seconds]
greaser|q has joined #osdev
[itchyjunk] has joined #osdev
* gog prr
<heat> pr
<moon-child> p
<heat> are you pushin p
<mcrod> wow
<mcrod> this is amazing
<mcrod> on windows now, when hiding the taskbar you no longer see a little sliver of it
<mcrod> this is... perfect now
<mcrod> almost perfect
<gog> nice
<gog> i always want to see the taskbar i have so many tasks
<heat> tasc
<heat> tascc
<mcrod> no you don't always want to see the taskbar
<mcrod> i hate taskbars, hate them
<mcrod> same with desktop icons
<heat> do you like GNOME
<mcrod> yes
<mcrod> modern GNOME
<mcrod> with the proper extensions
<mcrod> without them it is cancerous
<gog> ok
gioyik has quit [Quit: WeeChat 3.8]
<heat> vincent van gog
<mcrod> that's amazing.
<heat> kind of an overreaction it's not that funny
<gog> hahahahahahahahaha
* gog cuts off her ear
<heat> hahahaha now that's funny
<gog> hahahaha
<heat> send me a pic of your ear cut off
<gog> i'm too tired tonight
<gog> i'll do it tomorrow
<heat> this is WEAK energy
FreeFull has quit []
sortie has quit [Server closed connection]
sortie has joined #osdev
freakazoid332 has joined #osdev
xenos1984 has quit [Quit: Leaving.]
gog has quit [Ping timeout: 240 seconds]
gruetzkopf has quit [Server closed connection]
gruetzkopf has joined #osdev
riverdc has quit [Server closed connection]
riverdc has joined #osdev
[itchyjunk] has quit [Remote host closed the connection]
graphitemaster has quit [Server closed connection]
graphitemaster has joined #osdev
xenos1984 has joined #osdev
bradd has quit [Ping timeout: 240 seconds]
heat has quit [Ping timeout: 258 seconds]
bradd has joined #osdev
slidercrank has joined #osdev
gorgonical has left #osdev [Killed buffer]
gorgonical has joined #osdev
osmten has joined #osdev
Vercas has quit [Quit: buh bye]
Vercas has joined #osdev
Vercas has quit [Client Quit]
Vercas has joined #osdev
Arthuria has quit [Ping timeout: 240 seconds]
Vercas has quit [Client Quit]
Vercas has joined #osdev
Vercas has quit [Client Quit]
Vercas has joined #osdev
Vercas has quit [Client Quit]
Vercas has joined #osdev
Vercas has quit [Client Quit]
Vercas has joined #osdev
slidercrank has quit [Ping timeout: 240 seconds]
Vercas has quit [Client Quit]
Vercas has joined #osdev
toastloop has quit [Read error: Connection reset by peer]
elastic_dog has quit [Ping timeout: 240 seconds]
<gorgonical> In the context of my baremetal forth interpreter I have a problem that I want access to parts of the underlying ISA. What's the right way to do this: 1) write assembly words that use those and expose those as regular words, 2) write the words in forth and do some definition hacking to directly compile the machine code bytes into the defintion, thus eliding the need to reassemble the underlying interpreter, or 3) som
<gorgonical> ething else that I haven't thought through
elastic_dog has joined #osdev
mcrod has quit [Server closed connection]
mcrod has joined #osdev
slidercrank has joined #osdev
benlyn has quit [Ping timeout: 252 seconds]
benlyn has joined #osdev
kneskade has joined #osdev
benlyn has quit [Ping timeout: 240 seconds]
bliminse has quit [Quit: leaving]
GeDaMo has joined #osdev
goliath has joined #osdev
bliminse has joined #osdev
SGautam has joined #osdev
zxrom has quit [Quit: Leaving]
manawyrm has quit [Quit: Read error: 2.99792458 x 10^8 meters/second (Excessive speed of light)]
manawyrm has joined #osdev
zxrom has joined #osdev
lechner has joined #osdev
<moon-child> gorgonical: I believe 1 and 2 have both been done
eroux has joined #osdev
<sham1> IIRC 1 is more traditional for FORTH
<Ermine> FOOOOOORTH
<GeDaMo> Forth! :P
<GeDaMo> Damnit, did I miss a convo about Forth? :P
<Ermine> I just saw it mentioned
<GeDaMo> Me too
<Ermine> So I shitposted 'FOOOOOORTH'
<moon-child> FORTHFORTHFORTHFORTHFORTHFORTHFORTHFORTHFORTHFORTHFORTHFORTHFORTH STAAAAAAAAAAACKS
danilogondolfo has joined #osdev
rnicholl1 has quit [Quit: My laptop has gone to sleep.]
<Ermine> I guess it is not ok to compare pointers in C?
<Mutabah> depends on what you use that comparison for
<mjg> it is UB for arbitrary pointers
<mjg> but it works in practice bcause everyone is doing it
<Ermine> Ah, my case is 'from the same allocation', so should work'
<mjg> or you can cast to uintptr_t
<mjg> that is indeed safe
<Ermine> thank you
<sham1> Yeah, so for example with linker symbols like `extern int start_of_something[]` and `extern int end_of_something[]` it is indeed undefined behaviour to compare the pointers to those, but in practice it's fine
<sham1> Replace with arrays to unsigned char if using ints there makes you nervous, as it should
Left_Turn has joined #osdev
CommH has joined #osdev
nanovad has quit [Ping timeout: 240 seconds]
nanovad has joined #osdev
CommH has left #osdev [#osdev]
<kof123> the explanation i have seen is segmented pointers (as an example), did noone think to write a normalize_pointer() ?
<sham1> That's expensive
<kof123> lwayer: is it heavy? kid who smells: yes jurassic park lawyer: then put it down, it's probably expensive
<kof123> *lawyer
<sham1> Also how would you compare pointers in something like a Harvard architecture where a pointer could either be to data memory or program memory
<sham1> That's also the reason for why function pointers are incompatible with data pointers
gog has joined #osdev
<kof123> yes, but that is a separate issue :D
<sham1> Not really
<kof123> you are either comparing 2 data pointers, or 2 function pointers, or one of each
<kof123> so 3 cases there
<sham1> Well you still have situations where you might for example store constant data into progmem
<sham1> Kind of like how one uses .rodata
<sham1> That's a data pointer but to program memory
<kof123> sure, but then would it be a data pointer or function pointer?
<kof123> "for C"
<sham1> It's a data pointer to C, but it really cannot be compared to "data memory" pointers which in turn makes the comparison UB
<kof123> anyway i do not dispute the value of knowing such things, rather, how many people worried actually tested where it would matter?
<gog> hi
<sham1> Well it matters if one gets problems with doing this stuff, it might be the compiler being over-eager in optimising around the fact that it's UB
<kof123> that makes sense
<sham1> Like yeah, in practice the compiler will do the sensible thing, it doesn't literally summon demons from one's nostrils but it might just make assumptions that don't actually hold when executing
PapaFrog has quit [Quit: ZNC 1.8.2+deb2 - https://znc.in]
osmten has quit [Ping timeout: 246 seconds]
PapaFrog has joined #osdev
<mrvn> Arduinos have pointers to ram and flash.
<kof123> again -- what is that at the C level?
<kof123> if you go outside C data pointers and function pointers, then you have already gone outside C "guarantees"
<mrvn> kof123: and uint16 but you need different opcodes to access flash and ram.
<kof123> i actually agree though -- if you are on a platform with various pointer types, C won't save you, you better know what is actually going on
<sham1> That's the slight downside of C being a high level language
<sham1> Also upside in many cases tbf
<mrvn> The problem is that all (data) pointer types can be converted to void* or char* and then the compiler has no way to say the comparison is sensless because the type differs
<mrvn> If you have different pointer types then there should be different void* types.
q3lont has joined #osdev
<kof123> http://chameleon.synth.net/english/developers/ now you can test your 24/48-bit C code using the simulator included in the SDK, along with stdin and stdout. you're welcome! </nick burns>
<bslsk05> ​chameleon.synth.net: Home of the Chameleon
<kof123> it works with wine as well :D
osmten has joined #osdev
heat has joined #osdev
<gog> is there any way to make firefox yell at me if i'm trying to do a comparison on a function object because i keep forgetting ()
<gog> in javascript
<heat> yeah use typescript, more yelling in general
<sham1> Yeah, normal JS won't help you out
<Ermine> gog: may I pet you to compensate browser yelling at you
<sham1> I think that in this case gog wants the yelling
<heat> >browser yelling at someone because of borken js
<heat> haha very funny never going to happen
<heat> it'd be likelier for the browser to start autofixing your code and adding ()
<FireFly> y'know it'd be nice if they'd made that part of strict-mode but I guess not
<Ermine> mostly firefox yells about content security policy or something like this
<FireFly> but yeah your best bet is static analysis like typescript or whatever ig
<heat> browsers do be smoking that JS pack but you dare misconfigure CORS and they send a SWAT team to your house
<gog> it's not an option in this case, i'll just have to remember to call the function
<gog> i'm using react, not something that makes sense
nyah has joined #osdev
<heat> react doesn't support ts?
<gog> maybe it does idk
<heat> i know angular does
<gog> but i don't have time to learn yet another thing to make this easier i need to get it done soon
<gog> i have bug reports stacking up on me
<FireFly> yeah fair
<sham1> React supports typescript
<FireFly> valid
<gog> dang
<gog> wish i knew that before starting lol
<Ermine> Is it work or personal
<gog> work
<sham1> I get to work with years old jQuery and plain JS at work. !!FUN!!
<gog> same
<Ermine> How fun is it to support old IEs
<gog> and bit by bit trying to replace it all with more modern js
<gog> still glued to jquery for two applications. we use jquery-ui pervasively
<gog> but we're trying to replace the stuff that really doesn't need to be client-side with server-side
<gog> we should port the whole thing to Blazor with MVC
q3lont has quit [Quit: Leaving]
<heat> you really need java
<heat> java is great
<heat> server-side java, client-side java, java as an applet
<sham1> Server-side Java is unironically nice
<heat> noooooooooo
* Ermine drinks java
<FireFly> I dunno, keeping your servers on an island across the world seems inconvenient
<heat> the only thing i would ever use java for is an iceland volcano sacrifice
<sham1> Although I actually work at a project using the Vaadin framework, so the java is somewhere between being client-side and server-side. Client-side stuff directed from the backend. Fun!
<heat> JDK goes first, JRE does second
<sham1> And not only is it that framework, but an ancient version!
<Ermine> Hopefully it will trigger minecraft rewrite on something more suitable
<FireFly> heat: for hexing the interview? :p
<sham1> I love those posts
<sham1> It's so cursed yet so interesting and neat
* FireFly nods
<sham1> But yeah, professional Java development is interesting. And if I was working in another institution, I'd probably get paid nicely, but lol public sector wages
gmacd has joined #osdev
gmacd has quit [Ping timeout: 240 seconds]
gmacd has joined #osdev
<mjg> hey, at least in the public sector you have immense job security
<mjg> sup folks got some network equipment https://moeilijklastig.nl/verloopjes/
<bslsk05> ​moeilijklastig.nl: Rare verloopjes
gmacd has quit [Ping timeout: 240 seconds]
heat has quit [Remote host closed the connection]
corank has quit [Server closed connection]
heat has joined #osdev
corank has joined #osdev
heat has quit [Ping timeout: 240 seconds]
heat has joined #osdev
PapaFrog has quit [Quit: ZNC 1.8.2+deb2 - https://znc.in]
PapaFrog has joined #osdev
terminalpusher has joined #osdev
Maja_ has joined #osdev
heat has quit [Remote host closed the connection]
heat has joined #osdev
<mcrod> hi
<gog> mcrod: may i pet you
<mjg> maybe we should create #osdev-tinder or somethin'
<mjg> just sayin
heat has quit [Read error: Connection reset by peer]
heat has joined #osdev
<gog> mjg: may i pet you
<mjg> you can pat me on the back
* mjg preps to burp
<mcrod> gog yes
* mjg can't hold it anymore and burps
* mjg pets mcrod real quick
* mcrod prrs?
<mjg> be a man and burp
<mcrod> i do burp
<mjg> do you know how to do it on demand
<mjg> essential skill in elementaryu school in poland
<mjg> fwiw
* gog pet mcrod
<mcrod> yes
* mcrod prr
<mjg> aight
pg12_ has quit [Ping timeout: 240 seconds]
<gog> i cannot burp on command
<gog> never been able to
<Maja_> mjg: #osdev-tinder? IME that's just called the fediverse
Maja_ is now known as Maja
<mjg> gog: you et a pass
<mjg> it is a boy thing after all
<mjg> Maja: i don't know what fediverse is and i'm too afraid to google it
heat_ has joined #osdev
heat has quit [Read error: Connection reset by peer]
<GeDaMo> Distributed social networks
<gog> mastodon and such
<bslsk05> ​'Why are Kung Fu masters getting BEATEN UP in China?' by serpentza (00:23:07)
Terlisimo has quit [Remote host closed the connection]
Terlisimo has joined #osdev
Maja has quit [Ping timeout: 240 seconds]
<heat_> mjg, because they're dorks
<heat_> no "kung fu master" is cool, they're all dorks, all of em
<bslsk05> ​seclists.org: Full Disclosure: OpenBSD kernel relinking is not transactional and a localexploit exists
<gog> oops
<mjg> naah it will be fine
dude12312414 has joined #osdev
<puck> i mean, if you can modify the link kit you could easily just swap the kernel out entirely, no?
heat_ has quit [Remote host closed the connection]
<puck> right, and the hash verification in reorder_kernel is really only a guarantee that the user hasn't swapped out the kernel manually
<puck> (otherwise the relinking would undo your manual kernel replacement)
kazinsal has quit [Server closed connection]
kazinsal has joined #osdev
<zid> never heard of that subsystem before, sounds like it's runtime patching? or maybe kernel modules?
<puck> no, it literally just takes the .o files that the kernel is made of and relinks the kernel at boot
[itchyjunk] has joined #osdev
<zid> what, why
<puck> it's like ASLR but for the layout inside the kernel
<zid> Are they aware they're insane and trying to work on it, or is it just required?
<puck> it makes for less exploitability
<zid> And a shotgun cures cancer in a petri dish
<puck> also idk, relinking is relatively simple?
<HeTo> I think on the OpenBSD side they're looking at everyone that supports loadable kernel modules and thinking that they are crazy
<HeTo> well, just my conjecture
heat has joined #osdev
<heat> relinking is an ok security idea but i imagine it makes for a nightmare performance wise
<heat> do you have a fast kernel? who knows? roll the dice, literally
<mjg> "ey obsd do you havea fast kernel"
<mjg> they don't
<mjg> so your objection is rebutted
<heat> OPENBSD ROASTED BY LOCAL FREEBSD MAN
<mjg> they could maintain consistent performance profile despite relinking
<heat> "ey freebsd do you have a fast kernel"
<mjg> one thing you need to do is make per obj file data placement
<heat> "eh, kinda, its ok i guess"
<mjg> heat: depends, who is asking
<mjg> ey want to see the cringest shite ever
<heat> god said in the bible "thou shalt not make thou kernel 3 fast 5 u"
<mjg> bsd-related
<heat> yes
<Ermine> Happy birthday to freebsd
<mcrod> indeed
<mcrod> happy birthday
* mcrod party
<Ermine> obsd doesn't have kernel modules?
<heat> one less year til crippling death to freebsd
<mjg> Ermine: NOT SECURE
<heat> Ermine, no, openbsd people HATE usability
<bslsk05> ​twitter: <pr1ntf> FreeBSD: ␤ "Check out my horse power." ␤ ␤ OpenBSD: ␤ *seatbelt reminder warning constantly ringing* ␤ ␤ NetBSD: ␤ "We put a Chevy small block in a 2/3 scale P-51 WWII fighter. Let's see if it flies." ␤ ␤ DragonFlyBSD: ␤ "Matt's been in the garage for weeks but his flying car is awesome."
<nortti> Ermine: they removed it some years back, on the grounds that it is complexity where security issues might arise
<Ermine> Heck, MS-DOS feels more advanced and sophisticated
<mjg> shite bot
<mjg> lemme cp
<mjg> FreeBSD: "Check out my horse power."
<mjg> OpenBSD: *seatbelt reminder warning constantly ringing*
<mjg> NetBSD: "We put a Chevy small block in a 2/3 scale P-51 WWII fighter. Let's see if it flies."
<mjg> DragonFlyBSD: "Matt's been in the garage for weeks but his flying car is awesome."
<mjg> :d
<heat> did not disappoint, kinda cringe
<mjg> what the fuck is wrong with these people
<mjg> what do you mean kinda
<mjg> you fucking genz
<heat> i mean, it's not the cringiest thing i've ever seen
<mjg> fair
<zid> I mean, while furries exist, I think we have a life-chamption
<zid> champion*
* mjg is trying to think what to put in Onyx:
<heat> matt's been in the garage for weeks because he's too much of an asshole to work on the WWII fighter
<heat> mjg, onyx: garbage
<mjg> noo
<mjg> Onyx: "at least i don't have networking like sortix"
<heat> onyx: "yo mjg check out this SICK flamegraph"
<mjg> your flamegraphs do make me sick if that's what you meant
<heat> :(
<mjg> fucking PESSIMAL shit all over
<heat> don't worry i'll get actual QSBR RCU and then beat your ass
<mjg> people are slowing the kernel down faster than i'm unslowing it
<mjg> at this point all you have to do is wait
<heat> i plan to LGPL it so IBM doesn't take me to court
<mjg> i court ibm
<heat> maybe one day i'll grow some balls and communist my kernel into GPLv2
<Ermine> Well this is the most pessimal thing I've heard this year
<mjg> is emacs gplv2?
<mjg> basicall you should stick to what rms thinks is best
<heat> emacs is GPLv3-or-later
<Ermine> Da best
<heat> but GPLv3 is a farce
<mjg> now that i think about it the aboe is not that cringe
<mjg> FreeBSD: "Check out my horse power." -- it sucks
<heat> GPLv3 is the stalinism of GPL
<heat> whereas GPLv2 is the original marxist-leninist doctrine
<mjg> OpenBSD: *seatbelt reminder warning constantly ringing* you already have the seatbelt
goliath has quit [Quit: SIGSEGV]
<mjg> NetBSD: "We put a Chevy small block in a 2/3 scale P-51 WWII fighter. Let's see if it flies." it does not
<mjg> DragonFlyBSD: "Matt's been in the garage for weeks but his flying car is awesome." -- man LSD is the bets shit ever innit
<mjg> there you go
<heat> dragonflybsd: "matt's a cunt"
<mjg> i like the guy fwiw
<heat> the netbsd people didn't :(
<mjg> not that we always see eye-to-eye (in fact we mostly don't :S)
<mjg> i think you confused the guy with theo
<heat> OH SHIT YEAH
<heat> sorry matt
* mjg slaps heat
<heat> openbsd: "theo's a cunt"
<mjg> i don't like the guy fwiw
<Ermine> minix?
<heat> microkernal
<mjg> macrokneril
<heat> imagine participating in a microkernel vs monolithic debate in the 90s
<heat> virginest shit ever
<Ermine> It was fresh back then
<mjg> ey linus did you ever SNIFF a girl
<heat> where was bryan cantrill during those cringe ass discussions
<mjg> i'm guessing busy wanking over Sun's logo
<Ermine> But ppl engage in it even nowadays on moronix forums
<mjg> mate moronix is what you get if a tabloid had a comment sectoin
<heat> "vim best!!!!!!" "nooooooooo GNEW emacs best vim shit!!!!!!!!!!!!!!!!!"
<heat> me: microsoft word 03
<Ermine> mjg: yep
<bslsk05> ​imgflip.com: bell curve - Imgflip
<mjg> :X
<heat> mjg, btw thats not true, Sun engineers KISSED girls and FUCKED
<mjg> did you know one of the SUN ENGINEERS ran away with Bonwick's wife?
<mjg> serious
<mjg> so you do have a point
<heat> hahaha
<heat> the sun engineers put the d in dtrace
<mjg> aight enough with the locker room talk
<heat> sir this is irc
<heat> this is not a locker room why are you undressing yourself
<mjg> i'm calming down
<Ermine> btw heat, marxism-leninism is not original. Marx did the original, and Lenin made DLC
<heat> yeah, marxism being GPLv1 aka the shit one
<mjg> not true sco^Wcommunist
<Ermine> Gotta read both of them
<Matt|home> ..
<heat> mjg, you mofos ever got clang lock annotations in the kernel?
<mjg> no
<heat> hmm mkay
<heat> i noticed you got them in pthreads stuff
<mjg> this plausibly only works for userspace to begin with
<heat> why?
<mjg> shite, that's why
<heat> hm?
xenos1984 has quit [Quit: Leaving.]
<bslsk05> ​'Xu Xiaodong EXPOSES CHINA | Fake Martial Arts Masters DESTROYED' by TotallyPointlessTV (00:10:24)
heat_ has joined #osdev
heat has quit [Read error: Connection reset by peer]
<heat_> <heat> they work fine in the kernel, zircon has them all over
<heat_> <heat> i'm trying to make em work in mine but it's obviously not easy to backport out of the blue, lots of loosey-goosey locking
heat_ is now known as heat
<mjg> if they truly work that is something to possibly tkae note of
<mjg> frebsd has lolo annotations in comments and there was at least one project which tried to use them for formal verification
<heat> it looks kinda flawed but it seems better than nothing i guess
<heat> like, it kind of goes haywire when you start doing conditional locking
<heat> which yes, it's kind of like a code smell
<mjg> look mate im chillin here
<mjg> talking shop is for nerds
<mjg> and i aint one
<heat> it's for kung fu dorks
* mjg snorts fentanyl
<mjg> you have not watched he video mofer
pg12_ has joined #osdev
<heat> "ninjas are cool" - dorks
<mjg> turns out these supposed martial arts masters were like unix geezers
<mjg> little substance
<heat> literally
<heat> keith bostic is a black belt in capoeira
<mjg> there is also a contingent of self-proclaimed masters
<mjg> so there is that
<heat> >little substance
<heat> >self proclaimed masters
<heat> is this linkedin
<bslsk05> ​www.youtube.com <no title>
<nortti> "The Bizarre World of Fake Martial Arts"
<mjg> does that talk about EMPTY FORCE?
<mjg> that was funny af
<bslsk05> ​'EFO Empty Force FAIL' by Marc Mula (00:06:03)
<nortti> I think so, though it's been a while since I watched it
<mjg> it wants me to sign in to confirm age, which i' not doing
<heat> 30 min is too long for my genz brain
* mjg is secretly younger than heat
<nortti> drop it into mpv, and it should play just fine
<mjg> watch the 6 minute
<heat> can you sum it up into a 10 second tiktok
<mjg> ye
<mjg> people pretend dragon ball is real
<mjg> except beams are invisible
<heat> actually go to joe rogan's podcast and give him some clips, i'll watch those clips piece by piece over 2 months in youtube shorts
<nortti> and have for a good while
<mjg> falls apart when a "non-believer" volunteers to be attacked
<heat> "jamie bring up that dragon ball guy getting slammed by khabib on the octagon"
slidercrank has quit [Ping timeout: 258 seconds]
<mjg> burp
<mjg> you want some real shit, google for lightworkers
<heat> i'm a lightworker for the BSD community
<mjg> you clearly have not done it
<mjg> 's all i'm going to say
<heat> have not done what
<mjg> gogoled for what this shit is
<heat> i have
<heat> "The simplest way to describe lightworkers would be as beings who feel an enormous pull towards helping others. Also referred to as crystal babies, indigos, Earth angels and star seeds, these spiritual beings volunteer to act as a beacon for the Earth, and commit to serving humanity. " <-- me in the BSD community
<mjg> keep reading mofo
<mjg> you are missing the point
<heat> yup i'm all that baby
<heat> im one with the universe
<heat> the universe says "openbsd sucks" so that's my opinion
* mjg suspects heat is a lizard person
<heat> im a cloud person
<heat> CEASE YOUR INVESTIGATION
<heat> ok but actually real talk these people are fucking nuts haha
<mjg> you don't know fuck mat,e keep reading
<heat> i have read it all, even the comments
<mjg> bitch plz
<heat> "how do i find this innate power" and shit
<mjg> this is a part of a bigger ecosystem
<mjg> there is aliens around the planet
<heat> im gud person how do i find psychic
<mjg> testing if we are advanced enough
<mjg> 'n shit
<mjg> google for lightworker + roswell
<heat> hmm that reminds me of scientology
<mjg> you will find the alien who crashed
<heat> also a south park episode haha
<mjg> and is inow inhabiting a human body
<mjg> and o on
<mjg> php, node.js and other crap were totally trojan horses planted by aliens who wis h ill on humanity
<heat> jordan hubbard is a bad alien
<heat> linux torvald good alien, al viro bestest alian
<mjg> theo is a conflicnted alien
<heat> i like how even aliens recognize american exceptionalism
<heat> you won't find em visting bucovina, romania i'll tell you that
<mjg> ye they told me they wanted to do it, but then they watched "indepdence day" and they are going to go after the white house
<mjg> when the time comes
<mjg> for shits 'n giggles
Benjojo has quit [Server closed connection]
Benjojo has joined #osdev
<Ermine> How did discussion come to american exceptionalism
<heat> because aliens are only ever spotted there
gioyik has joined #osdev
<mjg> the rest of the world has cameras too good to spot aliens
<mjg> s/spot/record/
<Ermine> Haha
<Ermine> Well, only US has area 51
<gog> hi
<heat> gog, alien in iceland where?
slidercrank has joined #osdev
<gog> heat: probably the highlands
<Ermine> beneath the ground
<gog> it's a desert and it looks like the surface of the moon
<gog> no that's the huldufólk
<nortti> are you saying they are *not* aliens?
<gog> they're indigenous to iceland
<Ermine> gog: may I pet you
<gog> yes
* Ermine pets gog
<heat> motherfucker i've just realized
<gog> she's sad because i'm ignoring her
<heat> your vulcanos are probably full of aliens
<heat> xenu that bastard!
<gog> snæfellsjökull is where the journey to the center of the earth began
<heat> cool socks btw
<Ermine> Nice socks
<gog> :D thanks
<heat> Ermine, nice opinion
<Ermine> heat: cool opinion
<Ermine> btw
<mrvn> The only aliens in america are from mexico
<mjg> gog: your cat looks awful lot like a dog
<heat> no that looks exactly like a cat
<mrvn> mjg: god made people in his image but then he looked in a mirror and made cats
<mjg> maybe in area 51
<mjg> or the chernobyl zone
<heat> probably goes *bark bark* exactly like cats go
<Ermine> chernobyl is another kind of fun
<mrvn> the russians certainly had the better cover up for their ufo
xenos1984 has joined #osdev
<Ermine> mrvn: we have tunguska incident for this
<mrvn> Oh yeah, they had that too. SO was chenobyl their second ufo crash or did they screw up a test flight of a copied ship?
<Ermine> who knows...
slidercrank has quit [Read error: Connection reset by peer]
<bslsk05> ​gcc.gnu.org: Please, really, make `-masm=intel` the default for x86
<heat> please change the default into my preference, thanks
<Ermine> Lmao
<bslsk05> ​gcc.gnu.org: GNU = Junkware
<nortti> did y'all see how the ffmpeg project accidentally ddos'd GMP through github CI, leading to GMP blocking microsoft IPs?
<gog> lmao amazing
<bslsk05> ​gmplib.org: GMP servers are under DoS attack from Microsoft
<bslsk05> ​gmplib.org: Requests from Microsoft IP Addresses
<gog> also gcc generates broken intel assembly too
<gog> at&t syntax is ugly and unwieldly but gcc generates it more correctly than intel
<Ermine> I don't think that assembly syntax is a hill worthy to die on
<gog> maybe gcc should generate some kind of HLA syntax and have a third compiler that translates that
<gog> yo dawg we heard you like compilers
<puck> nortti: not the ffmpeg project
<nortti> oh, I thought it was ffmpeg you linked the commit to?
heat has quit [Remote host closed the connection]
<zid> gog when are we writing my compiler
<puck> nortti: just some guy running github ffmpeg builds https://github.com/BtbN/FFmpeg-Builds/commit/d75466340a9a5985e546fff9756b976727c4ab98
<bslsk05> ​github.com: Use old release version of gmp again · BtbN/FFmpeg-Builds@d754663 · GitHub
heat has joined #osdev
<nortti> ah, I see
gioyik has quit [Remote host closed the connection]
gioyik has joined #osdev
<gog> zid: i'm writing your compiler rn
<heat> nortti, this torbjorn is a bit of an idiot isn't he?
<gog> also i'm doing cleanup for my project now and so far i've deleted over 1600 lines of code
<gog> and i've only added 800-ish
<gog> it still has async quirks
<FireFly> heat: eh I can see the point sorta?
<FireFly> but he's definitely... opinionated
<heat> >guy abuses the shitty GMP server from github CI
<heat> hurr durr microsoft big tech bad
<heat> m$ moment
<Ermine> Isn't gmp part of gnu?
<nortti> yeah
<nortti> GNU Multiple Precision Arithmetic Library
<Ermine> Aren't they supposed to use gnu infrastructure (which is less shitty I guess?)
<gog> microsoft is bad
<Ermine> heat: maybe gmp people cannot afford github-scale infrastructure. Hence microsoft bad
<gog> let's not let a dude misconstruing ffmpeg abusing their git detract from this realty
<gog> it's not microsoft abusing the infra, it's ffmpeg
<heat> microsoft bad gnu poggers free as in freedom free as in beer
<zid> gog: oh great, what was the bugfix for line 4?
<gog> zid: delete it
<heat> (lets skip the huge contributions microsoft has made for open source)
<gog> yeah and i made a huge contribution to the toilet bowl this morning but it's the same thing
<Ermine> And almost closed source for some .net 6 features
<heat> Ermine, there's no need for github-scale infra, dude, a single guy fucking DDoSed their thing
<heat> "beefy" my ass
<nortti> < gog> let's not let a dude misconstruing ffmpeg abusing their git detract from this realty ← not git, mercurial. dunno how well the mercurial server side handles full clones vs. git
terminalpusher has quit [Ping timeout: 246 seconds]
<Ermine> My infra is a single 1G double core VPS
<puck> idk, not a lot of projects need to deal with 100-ish full clones in a short period, i'd say
<puck> (on mercurial)
<heat> GMP is a dependency for a whole lot of important projects
<gog> nortti: oh fuck
<gog> why are they using hg
<FireFly> good question
<nortti> I have no idea. at least they're not using bazar
<gog> i used bzr for awhile
<gog> then i got git and got gud
<nortti> but yeah, was fun figuring out how to contribute to it
rnicholl1 has joined #osdev
<FireFly> I wonder what he does these days
<FireFly> I mean outside of maintaining GMP
<gog> blames microsoft for various problems in his life
<FireFly> apparently yeah
<gog> "whoops, i missed the bus. goddamnf ucking microsoft"
xenos1984 has quit [Ping timeout: 258 seconds]
<FireFly> but he used to work at uni, but there was some drama and he quit in protest or so
<gog> "ah shit my food is burning, microsoft why didn't you turn off the oven when i asked you?"
<FireFly> hmm
<FireFly> wonder if KTH still has a compiler construction class, he used to run it
<FireFly> https://www.kth.se/student/kurser/kurs/DD2488/?l=en oh rip, getting discontinued
<bslsk05> ​www.kth.se: KTH | DD2488
<heat> the gcc mailing list is so toxic
<bslsk05> ​www.mail-archive.com: Will GCC eventually learn to use BSR or even TZCNT on AMD/Intel processors?
<Ermine> heat_: microsoft could rate limit requests to gmplib site as a fast remedy
goliath has joined #osdev
<heat> just to be clear, microsoft is not github and they're separate entities
<heat> 2) the load is not all that damn much for a freaking cornerstone in GCC. you need this to build GCC!!!
<heat> yes, 40-clone-burst is weird and kind of abusey. it should not however screw your whole infra
<heat> make them rate limit clones if they care
<heat> if a rando can take down their thing in github actions, so can I with a simple bash for loop
<Ermine> Better infra would be cool to have, but what if project doesn't have enough money for it?
<gog> what if we stopped having projects
<gog> no more code
<Ermine> MS/GH/whatever has loads of $$$. Open source projects don't necessarily have $$$. Gues who has to accomodate who?
<gog> fix the code we have
<heat> this does not necessarily need more infra
<heat> just incredibly basic iptables rules work
<Ermine> so those rules are in place rn
<heat> no they're not
<heat> they're blocking whole ranges of IPs, but you can throttle with iptables
<heat> is this going to be manual? what's stopping me from being a dick and doing the same?
xenos1984 has joined #osdev
<nortti> I think the problem is less the amount of data transferred (which they can handle) and more the CPU power required to compress the entire repository for transfer
<mcrod> wait did something happen at github
<nortti> in which case any rate limiting has to happen at a higher level that knows whether you're, say, downloading release tarballs (cheap) or doing full clones (expensive)
<heat> mcrod, some rando's CI script cloned GMP way too much, and GMP died because of it, then blamed le bad microsoft
<nortti> if you limit any traffic to the level where full clones would not be a problem, you're making everyone downloading tarballs' day worse for no real benefit
<mcrod> that's actually hilarious.
<puck> tbf the response from BtbN is really bad
<puck> > Their hg server is way too flakey to rely on, download a release tarball from GNU mirror instead.
<puck> > I hope no important fixes are missed due to the loss of update automation this causes.
<heat> nortti, considering clones are taking it to its knees, it may be a solid idea to throttle everyone in general
<heat> in any case, cloning sounds highly cacheable
<mcrod> god I wish clang-tidy would shut the fuck up sometimes
<heat> same
<heat> but you can just disable some checks
<gog> but you have a style violation
<gog> you can't commit a style violation that's cringe you're going to lose contributor
pg12_ has quit [Ping timeout: 252 seconds]
<heat> most of the modernize checks are really stupid and i hate them
<heat> like it wants me to replace every define with an enum
<gog> sometimes you can't
pg12_ has joined #osdev
<mcrod> yeah I've begun to disable a few of them
<gog> when is C getting constexpr
<mcrod> mostly the "use a trailing return type" ones
<mcrod> gog C2X
<gog> it's in c2x??
<mcrod> the keyword is, yes
<mcrod> most of the people in #c are having strokes over the 'new blood'
<heat> i have strokes over the C VLA syntax
<heat> char buf[static 8] or some garbage like this
<mcrod> doing that isn't even implemented in gcc
<mcrod> it's basically a nop
<heat> yeah they don't warn you on OOB
<nortti> I though it would give you warnings for NULLs at least
<gog> i used a VLA for one thing in the past and then realized that VLAs are really bad actually
<nortti> also I don't think that's a VLA, I think it's just a normal pointer
<heat> yes, it's a VLA-syntax-like thing
<heat> but a pointer
<mcrod> what the fuck kind of warning...
<mcrod> "member has public visibility"
<heat> you know what i also have strokes over? namespaced symbols
<mcrod> maybe it's fucking _supposed to_
<heat> stdc_ and posix_
<heat> stdbit.h should burn in hell
<gog> wow, my pharmacy has an app where i can order my refills
<gog> i no longer need to call and speak my broken icelandic
gog has quit [Quit: Konversation terminated!]
<heat> i actually dont have all that many shit clang-tidy warnings because i have an extremely-opinionated .clang-tidy
<heat> mcrod, https://github.com/heatd/Onyx/blob/master/.clang-tidy if u subscribe to my brand of c++
<bslsk05> ​github.com: Onyx/.clang-tidy at master · heatd/Onyx · GitHub
<sham1> What's your brand of C++
<GeDaMo> C++ requires you to get branded now? :|
<heat> no exceptions, no rtti, no const correctness, no always-auto, sometimes very C like, little to no templates
<Ermine> C with classes
<heat> yeah kinda
<heat> C with types
<Ermine> I'd like it actually
<Ermine> structs but they have constructors and destructors
<heat> join onyx
<nortti> do you use std::move in onyx?
<heat> yes
<sham1> good
* sham1 looks at readme
<sham1> > GNU plus Onyx
<nortti> ognyx
<mjg> i go out, come back, and you are still shitposting
<mjg> you absolute legends
<heat> we're absolute nolif-yes yes legends haha we're legends
<sham1> Those aren't mutually exclusive
<Ermine> nortti: ognux
<mjg> debian gnu/heat
<mjg> did you know there are peeps who try to make darwin-xnu repo into a full os?
<heat> yes
<mjg> that is both better and worse than ractos
<nortti> Ermine: ognуx, with cyrillic у
<heat> also I don't associate with debian, sorry
<Ermine> nortti: ah, lmao
<sham1> heat: what do you use? Arch?
<heat> yes btw
<sham1> CALLED IT!
<mjg> did you mean gnu arch?
<heat> mjg uses gnu hurd
<mjg> no, i use plan9
<mjg> not gnu!
<sham1> Oh yeah, forgot. Clearly has to be GNU Linux-Libre Parabola
<heat> usually i ask for a cuba libre but i can try a gnu linux-libre next time i go for a drink
<mjg> Download ReactOS 0.4.14 or Donate
<mjg> between the two i chose Donate
<sham1> I've actually unironically used Parabola Linux-Libre unironically because it has actually worked with some of my systems better than distros with proprietary packages (shock horror). It's been fixed IIRC but yeah
Arthuria has joined #osdev
<sham1> > mfs say the same thing twice because you forget what you just wrote
<mjg> you mean neither work now? :rimshot:
<sham1> Hah!
<mjg> why on earth is it so hard to find a short-sleeve shirt with a pocket
<heat> why do you want a shirt with a pocket
<sham1> No one wants to make nerd clothing
<mjg> look genz it is too complicatd for you to understand
<heat> it looks stupid and does nothing except maybe hold a pen
<mjg> i use a 3 pocket system
<mjg> so
<mjg> FU
<sham1> I'm sure he'd also go for pocket protectors
<mjg> if they were not so mainstream, maybe
xenos1984 has quit [Ping timeout: 260 seconds]
gog has joined #osdev
slidercrank has joined #osdev
xenos1984 has joined #osdev
justmatt8 has quit [Server closed connection]
justmatt8 has joined #osdev
<mcrod> god. i forgot you can't use memcpy in a constexpr function.
<kof123> " fix the code we have"
<kof123> you just tanked the stock market gog
<kof123> so...if you short i guess it would work
<kof123> otherwise, not happening
<gorgonical> re: my forth interpreter. I think both are permissible. The jonesforth uses option 2 and manually builds the machine code in the definition by writing it byte-by-byte. That gets you the "spirit of minimalism" that the interpreter doesn't do anything in asm other than what's needed to run the forth itself, and "fancy" things like being able to use the fence insn are left up to you
<gorgonical> But whereas writing inline asm in C feels like a convenience, writing machine code in forth words feels like the opposite lol
<gorgonical> Something about bypassing the assembler and utilizing raw machine code feels absurd when the .S file I wrote makes extensive use of it anyway. Not just labels and b 1000f or whatever but descriptive labels like end_of_loop
<heat> noooooo
<heat> you probably want .Lend_of_loop instead
<heat> .L labels do not create a local symbol, end_of_loop: will
<gorgonical> what's a local symbol in this context
<heat> ELF symbols
<gorgonical> good news: its one file
<heat> doesn't matter, if you look at the readelf you'll probably have a boatload of extra symbols
<gorgonical> No extra symbols in my readelf output
<heat> hmm ok then
<heat> that's odd
<gorgonical> In fact it looks like the assembler got rid of all the labels internally. Only .globls are there
<gorgonical> Oh no that's not true, the loop labels are there
<gorgonical> I missed them at first
osmten has quit [Quit: Client closed]
<gorgonical> what's this stray file symbol in my symtab? ccrkNIl0.o ? Its value is zero, is that some c runtime stuff?
<nikolar> sounds like it
<kof123> looks more like a temporary filename IMO
<heat> i haven't seen .o but i have seen plenty of .c
<kof123> ^ i would expect a source filename, not a .o
<nikolar> nothing shows up when you look it up so guess it's a temporary
<gorgonical> I wonder how it got in there
<gorgonical> indeed if I rebuild it changes to cclFeVPe.o so it's some kind of temporary
<gorgonical> Some compiler temporary I'd guess
<kof123> if you run -v i think gcc spits out that stuff...dunno about clang/etc.
<gorgonical> I can see now it's a temp arg passed to collect2
<gorgonical> -static /tmp/ccMrASSm.o blah blah
<zid> image not -pipe
FreeFull has joined #osdev
<zid> Do you think factorio is compiled with -pipe
<zid> Or is it compiled with -belt
<gorgonical> You beat me to the joke
vdamewood has joined #osdev
gmacd has joined #osdev
<zid> 2 hours for honzuki
<zid> <matja> --no-relax for deathworld
rnicholl1 has quit [Quit: My laptop has gone to sleep.]
rnicholl1 has joined #osdev
SGautam has quit [Quit: Connection closed for inactivity]
vdamewood has quit [Quit: Life beckons]
sortie has quit [Read error: Connection reset by peer]
sortie has joined #osdev
rnicholl1 has quit [Quit: My laptop has gone to sleep.]
rnicholl1 has joined #osdev
bas1l has joined #osdev
exec64 has quit [Read error: Connection reset by peer]
utzig has joined #osdev
listentolist has joined #osdev
exec64 has joined #osdev
basil has quit [Ping timeout: 265 seconds]
drakonis has quit [Ping timeout: 240 seconds]
drakonis has joined #osdev
GeDaMo has quit [Quit: That's it, you people have stood in my way long enough! I'm going to clown college!]
Turn_Left has joined #osdev
Left_Turn has quit [Ping timeout: 265 seconds]
goliath has quit [Quit: SIGSEGV]
DoubleJ has quit [Server closed connection]
DoubleJ has joined #osdev
gmacd has quit [Ping timeout: 265 seconds]
gmacd has joined #osdev
gmacd has quit [Ping timeout: 240 seconds]
danilogondolfo has quit [Remote host closed the connection]
gmacd has joined #osdev
rnicholl1 has quit [Quit: My laptop has gone to sleep.]
ThinkT510 has quit [Quit: WeeChat 3.8]
zxrom_ has joined #osdev
zxrom has quit [Ping timeout: 265 seconds]
eddof13 has joined #osdev
ThinkT510 has joined #osdev
varad has quit [Server closed connection]
varad has joined #osdev
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
gmacd has quit [Ping timeout: 240 seconds]
gmacd has joined #osdev
<heat> strictly speaking the operating systems with the strongest UNIX roots are atm illumos, HPUX and AIX
<heat> hence superior to the dirty separatist BSDs
<nortti> what about SCO openserver?
<heat> oh that's a good one too!
<heat> i think there's also a unixware version these days, somewhere
zxrom_ is now known as zxrom
<kof123> where is plan9 in this equation?
<kof123> there's roots and then there's Roots
<heat> plan9 is also a spinoff, but an even worse one than BSD
<heat> probably THE worst spinoff
<heat> and not really that spin off either
greaser|q has quit [Remote host closed the connection]
gmacd has quit [Ping timeout: 240 seconds]
<heat> probably something like illumos (maintained, open source) > AIX > HPUX > SCO openserver (really not maintained much) > BSD (spun off, but has some good UNIX heritage) > macOS Darwin (wtf is going on here) > Linux (dirty nordic UNIX) > plan9 (microkernel)
<kof123> i would wild guess digital unix/tru 64 is in bsd/darwin/machland there
<kof123> *in between
<nortti> how does 2.11BSD rank? still mostly based on unix v6 from what I understand, open source, and still technically getting patches about once a year
<bslsk05> ​www.tuhs.org: Index of /Archive/Distributions/UCB/2.11BSD/Patches
<nortti> also quite a few in 2022. guess 2021 was just a slow year for it
gmacd has joined #osdev
<heat> hmm, i mean, it's a BSD, but still pretty early so very UNIX-like (at some point)
<heat> maybe more UNIX than BSD but less UNIX than openserver/unixware
<heat> certainly cannot overtake any of the proprietary UNIXes and their beautiful amazing code and idea sharing with AT&T/USL itself
gmacd has quit [Ping timeout: 240 seconds]
gmacd has joined #osdev
<Ermine> plan9 is not microkernel afaik. At least its authors didn't consider it as such
* kof123 wipes dust off monitors for rare daytime usage "whoah" </keanu>
vin has quit [Remote host closed the connection]
gmacd has quit [Ping timeout: 240 seconds]
heat has quit [Read error: Connection reset by peer]
heat_ has joined #osdev
rnicholl1 has joined #osdev
gioyik has quit [Quit: WeeChat 3.8]
gmacd has joined #osdev
freakazoid332 has quit [Ping timeout: 260 seconds]
heat_ has quit [Read error: Connection reset by peer]
heat_ has joined #osdev
<zid> how are we, my sweaty friends
vin has joined #osdev
gmacd has quit [Ping timeout: 240 seconds]
gmacd has joined #osdev
<mcrod> i'm dying
kneskade has quit [Ping timeout: 240 seconds]
gmacd has quit [Ping timeout: 258 seconds]
Archer has quit [Server closed connection]
rein-er has joined #osdev
<bl4ckb0ne> hi dying im dad
mrvn has quit [Ping timeout: 258 seconds]
<mcrod> from now on you're not known as bl4ckb0ne
<mcrod> you're known as meson
<bl4ckb0ne> no im dad
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
heat_ has quit [Remote host closed the connection]
heat_ has joined #osdev
Reinhilde has quit [Server closed connection]
Arthuria has quit [Ping timeout: 240 seconds]
<Ermine> bl4ckb0ne dad
<bl4ckb0ne> i can call you sports and ask you if you're winning
doppler has quit [Server closed connection]
doppler has joined #osdev
Ellenor has joined #osdev
freakazoid332 has joined #osdev