<bslsk05>
homepage.divms.uiowa.edu: Jones on reciprocal multiplication
gildasio1 has joined #osdev
<kof123>
quote there are some significant differences between ANSI C (C89) and the dialect recognized by compilers supported by EDK II (C95)
<heat>
intel lives in an alternative dimension where C95 was a thing
<zid>
oh look there's the shift, neat
<heat>
actually, c95 was a thing, wow
<kof123>
i actually did the avoid initializers, but that was for $CRAZY_META_REASONS i am just curious if you are really doing that heat, or it is part of hard mode aoc
<zid>
If you could make me a lookup took where I give it a constant and a shift amount
<zid>
I'd actually find that useful
<heat>
kof123, hard mode k&r aoc
<heat>
it looks partially like a shell script and I'm very happy with that
<heat>
I could've omitted a few ints here or there, I need to get better at this
<bslsk05>
github.com: aoc2022/day1.krk at master · klange/aoc2022 · GitHub
<zid>
what's kirk
<moon-child>
main dude from star trek
<zid>
Makes sense
<zid>
nerds love naming things after star battle show
<gog>
zid is quark
<zid>
quirk
<zid>
our bastard lovechild
<klange>
While I admittedly read the file extension as 'kirk', the language is Kuroko.
<zid>
yea it's your python thingy right
<zid>
that you write editors in
<kazinsal>
gonna start a fork called stackless kuroko
<moon-child>
using exclusively cps?
<klange>
The syntax highlighters and theme files are written in Kuroko, the editor itself is C.
epony has joined #osdev
<klange>
kazinsal: kuroko already has the primary feature of stackless python, which isn't stackless, it just avoids C call stacks for managed function calls - Kuroko already does this in many cases
<heat>
you didn't pick a license for your aoc solutions
<heat>
i'm gonna steal all your code
<heat>
and then.. uh.. i'll
<kazinsal>
implicit proprietary code
<klange>
Without a license, being on github grants certain rights implicitly, such as your ability to read and fork it, but, since you mention it, I'll slap the ISC on the whole thing.
<heat>
you should GPLv2 it like me
<heat>
don't like the big corps steal your aoc solutions!
<heat>
kazinsal, i'm entirely positive code licensing as is currently is very iffy and based on bogus
<heat>
as in, nothing makes sense and very few of it seems to be enforceable
<kazinsal>
code licensing law is entirely based on who has better lawyers
<heat>
it's like the blatant GPLv2 violations in china and shit
<heat>
or if reading code for documentation makes your work a derivative work of that
<heat>
and as far as I understand it's a don't-know-don't-ask
<kazinsal>
a few years ago there was that whole thing about the gpl being enforceable as a contract which is funny as hell considering the FSF goes "NO IT'S NOT" despite that if the courts went "okay, have it your way, it's not" it'd be unenforceable
<kazinsal>
but in reality the enforceability of something as abstract as a copyleft viral software license is based on who has a bigger lawyer budget
<heat>
tldr nothing makes sense we're all fucked
<kazinsal>
it would be equal parts horrifying and utterly popcorn-level fascinating to watch the FSF try to take on, say, Oracle
<heat>
i think that the big corps would ironically back the FSF
<heat>
linux is only useful while people and corps keep contributing
gildasio1 has quit [Remote host closed the connection]
gildasio1 has joined #osdev
FreeFull has quit []
Burgundy has quit [Ping timeout: 252 seconds]
gog has quit [Quit: byee]
<epony>
let's call it UNIX downstream experimental branches
<epony>
or Linus' coordinated collaborative semi-compatible collection
<geist>
heat: cARGOCOGOARCOGO
<epony>
if it's not this it would be that, and still there would be a system like UNIX for consumer-corporate shared work
<epony>
it isn't everything and not the only one, just one of a couple more "public" and "general purpose" yet experimental and non-standard / reliable long term
gog has joined #osdev
<epony>
do the account for zipbombs ;-)
<epony>
"they"
<geist>
huh lamborghini is apparently announcing an offroad rally supercar
<zid>
a.. rally.. supercar?
<zid>
or do they just mean a "very good rally car"
<GeDaMo>
«The current spelling, from the 16th century, is due to association with Medieval Latin scissor (“tailor”), from Latin carrying the meaning “carver, cutter”, from scindere (“to split”).» https://en.wiktionary.org/wiki/scissors
<bslsk05>
github.com: aoc2022/day2.c at main · heatd/aoc2022 · GitHub
<zid>
You can do it in one modulo and an add I think, btw
xenos1984 has quit [Read error: Connection reset by peer]
<zid>
from my xor fuckery it seemed that way at least, I bet the reddit thread has it
<heat>
I find it mildly annoying how modern compilers can't take "i, j;" as "int i, j;"
<zid>
without the type it could be a comma operator expression I guess?
<heat>
and no I'm not installing gcc 2.91 or whatever the last fully K&R supporting compiler was
<heat>
even without the comma, "i;" isn't a valid decl or i
<heat>
of*
<heat>
hmmmmmmm
<heat>
I see, I think I'm missing the auto
<heat>
"auto i, j, k;" defines 3 ints
<zid>
(void)i; is the one true expression
MarchHare has joined #osdev
<heat>
the register and auto keywords are the true expressions all by themselves
CryptoDavid has joined #osdev
<zid>
expression of your inability to write good, proper, modern, C89
<heat>
fun fact it's not legal to take the address of a register var
<zid>
hardly a fact
<heat>
your mum's hardly a fact
<zid>
more of a "wtf would the result even be?"
<zid>
Did you know you can't take the inverse upper row echelon form of a dog
<heat>
well in a modern compiler you would spill it to the stack and that would be it
<zid>
"Well in a modern compiler you'd just take the FFT of the dog"
<heat>
but register is for the olden days
<heat>
i wonder if register does indeed affect any codegen in gcc/clang
<zid>
I'd be surprised, but if the answer is yes I wanna hear what
<heat>
like, if it needs to spill a register to the stack, does it prefer to spill non-register over register? that would make /some sense/
<heat>
non-register vars over register vars*
<heat>
or maybe it just wouldn't spill a register var, ever
<heat>
because you need it to be in a register ofc!
<heat>
else why would you add that keyword
xenos1984 has joined #osdev
hextuff has joined #osdev
<heat>
he implementation may treat any register declaration simply as an auto declaration. However,
<heat>
whether or not addressable storage is actually used, the address of any part of an object declared with
<heat>
storage-class specifier register cannot be computed, either explicitly (by use of the unary &
<heat>
operator as discussed in 6.5.3.2) or implicitly (by converting an array name to a pointer as discussed in
<heat>
6.3.2.1).
<heat>
zid, how does your bithacking shittery work?
<zid>
idk, I made a truth table
<zid>
I needed to turn 012 into 201 or something
<zid>
so I did some xors and a subtraction
netbsduser has joined #osdev
<zid>
I could remove the == with some more xors ofc
<zid>
== is actually the xnor operator
gildasio1 has quit [Ping timeout: 255 seconds]
gildasio1 has joined #osdev
gildasio1 has quit [Remote host closed the connection]
gildasio1 has joined #osdev
bgs has joined #osdev
bgs has quit [Remote host closed the connection]
joe9 has quit [Quit: Lost terminal]
gildasio1 has quit [Remote host closed the connection]
joe9 has joined #osdev
gildasio1 has joined #osdev
Piraty_ is now known as Piraty
heat_ has joined #osdev
heat has quit [Read error: Connection reset by peer]
heat_ is now known as heat
MarchHare has quit [Ping timeout: 256 seconds]
MarchHare has joined #osdev
gildasio1 has quit [Ping timeout: 255 seconds]
gildasio1 has joined #osdev
Burgundy has quit [Ping timeout: 260 seconds]
rty has joined #osdev
potash has joined #osdev
rty has quit [Quit: rty]
ss4 has joined #osdev
wootehfoot has quit [Ping timeout: 260 seconds]
Burgundy has joined #osdev
rwxr-xr-x has joined #osdev
terminalpusher has quit [Remote host closed the connection]
terminalpusher has joined #osdev
ss4 has quit [Quit: Leaving]
wootehfoot has joined #osdev
terminalpusher has quit [Remote host closed the connection]
mykernel has quit [Read error: Connection reset by peer]
mykernel has joined #osdev
CryptoDavid has quit [Quit: Connection closed for inactivity]
Matt|home has joined #osdev
FireFly has quit [Ping timeout: 255 seconds]
FireFly has joined #osdev
hussein has joined #osdev
nur has quit [Ping timeout: 265 seconds]
carbonfiber has joined #osdev
xenos1984 has quit [Read error: Connection reset by peer]
<rwxr-xr-x>
I have a very strange problem
<rwxr-xr-x>
I use WSL2 right? Trying to run anything in the terminal that opens ports or connections fails
<rwxr-xr-x>
for example, gdb targeting remote
<rwxr-xr-x>
times out
<sbalmos>
host Windows firewall
<rwxr-xr-x>
I checked, it's fully let through
<sbalmos>
WSL guest firewall
Dyskos has joined #osdev
xenos1984 has joined #osdev
<rwxr-xr-x>
let me check that
<rwxr-xr-x>
if that doesn'twork, i'm gonna try shutting off firewall
xenos1984 has quit [Ping timeout: 256 seconds]
<rwxr-xr-x>
Both didn't work lol
<rwxr-xr-x>
Firewall off everywhere
<rwxr-xr-x>
kali and windows
xenos1984 has joined #osdev
<heat>
>kali
<heat>
there's your issue, hackers get no internet
<rwxr-xr-x>
so true
<rwxr-xr-x>
i have like 6 vulnerable web apps when I get bored, on my localhost lol
wootehfoot has quit [Read error: Connection reset by peer]
<epony>
wipe windows clean
<rwxr-xr-x>
lmaoo
<epony>
too much bugs on the windshield obstruct visibility
<epony>
there are cracks in it in a lot of places too
<epony>
and hole punchin is two layers of windows glassware with glue in between
<epony>
compat layers are extra trouble and dependence on the host system which has defects and unreliable / unstable defects
<epony>
it's not a well designed system in general
<epony>
the advertisements stickers all over the windows does not improve visibility and sight
<epony>
also it's pluggable, so libraries replacements and hooks by "security" software add one more layer to validate
<epony>
it's better to use a real virtualisation hypervisor than the built-in shim
Starfoxxes has joined #osdev
<rwxr-xr-x>
Okay, so i'm trying to debug my program with gdb, and i'm starting to suspect that things may be more broken that it seems.
<mjg>
you should valgrind it first
<rwxr-xr-x>
its says it loads the symbold from the kernel.elf file, but it sure doesn't seem like it does, it shows all of them whenever I tab and print, but it always says it cannot access memory address (ADDRESS)
<rwxr-xr-x>
valgrind?
<rwxr-xr-x>
looks cool
<rwxr-xr-x>
let me figure out how to use it
rpnx_ has joined #osdev
vdamewood has quit [Quit: Life beckons]
invalidopcode has quit [Remote host closed the connection]
invalidopcode has joined #osdev
epony has quit [Quit: QUIT]
xenos1984 has quit [Ping timeout: 252 seconds]
rpnx_ has quit [Quit: Leaving]
xenos1984 has joined #osdev
poisone has joined #osdev
rwxr-xr-x has quit [Ping timeout: 252 seconds]
Terlisimo has quit [Quit: Connection reset by beer]
smach has quit []
Terlisimo has joined #osdev
<vai>
coffee rules
mykernel has quit [Ping timeout: 255 seconds]
gog has quit [Read error: Connection reset by peer]
elastic_dog has quit [Read error: Connection reset by peer]
elastic_dog has joined #osdev
<heat>
you can't use valgrind there
<heat>
sorry
gog has joined #osdev
<sbalmos>
can't use valgrind with coffee? wtf?
<sbalmos>
how am I supposed to grind the beans?
<geist>
coffffeeeeeeee
<geist>
what everyon needs is a CRATE of coffee
<geist>
i like my coffee to have a RUST like color
<heat>
CRATECREATE CARGOOOOOOOOOOO
<heat>
port zircon to riscv64? if it were written in rust you could just import the riscv64 crate
<heat>
rust: come for the memory safety, stay for the licensing-ambiguous crate system
<heat>
in fact, zircon? just import the linux-kernel crate lol wtf
<sbalmos>
yeah sometimes the crate craze is a bit as bad as npm-hell
<sbalmos>
I don't want to rely on the x86-64 or uefi crates as I'm starting out here. too much hiding.
<heat>
you can write uefi applications using the full range of rust std
<heat>
feel the powah of rust1!!11111111!!!!111111111
<j`ey>
sbalmos: youre probably just going to end up with your own versions of those crates in some form
<heat>
if you're borrowing architecture support you might as well just cp -r freebsd-src myos
<sbalmos>
j`ey: true, in classic NIH fashion. but it's for the process of learning and not having a magic "oh just have this entrypoint method and this attribute or this uefi_entry!() macro call"... "but... the UEFI spec says the entryp..." "USE THIS MACRO! STOP WASTING TIME!"
<heat>
it's not like having very modular crates is a good thing
<heat>
how can you properly integrate code into your system if there's a very strict boundary you can't cross, improve or further integrate with your system
<heat>
at least the good old tried-and-true method of copying freebsd code works because you can literally *change the code* and adapt it
<heat>
and you also get to control the actual code and not have a dubious crate supply chain powered by licensing ambiguity where, as far as I understand, everyone gives 0 fucks and assumes crates are apache-2.0 licensed or whatever
<sbalmos>
my fucks are on backorder
<heat>
"god dang good ol days with the good ol dang C and 'em manuals and man pages, not 'em fancy god dang rust crates and npm jibber jabber"
dude12312414 has joined #osdev
<sbalmos>
okay Boomhauer
bauen1 has quit [Ping timeout: 256 seconds]
bauen1 has joined #osdev
mykernel has joined #osdev
<mrvn>
licenses could be sort of dependencies in crates
<mrvn>
and beware: With GPL you can only use crates supplied by your OS.
<mrvn>
(because of the OS exception in the GPL allowing you to mix licenses)
<sbalmos>
hmm... crates should declare its license and allowable dependent licenses?
<j`ey>
it declares the license, but i dont think you can do the latter
<sbalmos>
cargo itself would then have to know a matrix of what licenses are compatible with what other licenses, so it could warn or stop if you're including a dep with an incompatible license
poisone has quit [Read error: Connection reset by peer]
<j`ey>
I think there are tools that do it for you, or at least gather all the licenses
<mrvn>
sbalmos: and now you need a formalized grammar to express license terms and a solver for them.
<sbalmos>
no terms! no surrender!
<mrvn>
sbalmos: License: You may link with a crate licensed under A but only if there is also a crate licensed under license B. Now that will be fun.
<sbalmos>
yeah, uhm, no
<sbalmos>
where's the attorney crate?
<mrvn>
that was the first crate the destroyed when the revolution begun
<mrvn>
+y
mykernel has quit [Ping timeout: 268 seconds]
mykernel 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!]
jafarlihi has joined #osdev
<jafarlihi>
Is there any reason to use ncurses in 2022? Seeing that btop rolls their own and looks much better than anything I have seen in ncurses, is the point about terminal compabilitiy moot now?
epony has joined #osdev
poisone has joined #osdev
Dyskos has quit [Quit: Leaving]
gareppa has quit [Quit: Leaving]
<geist>
not sure what btop is, but at the minimum it means you dont have to roll your own either
jafarlihi has quit [Ping timeout: 260 seconds]
<particleflux>
Does btop use their own stuff for the UI?
<particleflux>
It's hard enough to be cross terminal emulator compatible, yet alone support running things on a Mac *yuck*
<geist>
not sure why the mac termainal app would be any different
<particleflux>
eh, I'm maybe thinking too much in terms of scripts
<particleflux>
mac with their stoneage bash, openssl, etc
carbonfiber has quit [Quit: Connection closed for inactivity]
<particleflux>
unless you use 3rd party homebrew
<geist>
yah fair i always use something like homebrew or macports so kinda forget what the stock experience is
<particleflux>
I created several scripts on linux, thinking they run on mac, and it always ended it some kind of disaster lol
<geist>
sometimes that's just BSD vs linux. happens a lot
<particleflux>
like, you have that old bash version, so \e ANSI escapes won't work
<particleflux>
and then, tehy also don't have newer openssl, so newer openssl cmds don't work either (attempted to use them for auto-converting appstore certificates for push msgs)
<mjg>
i don't think the terminal wars are dead
<particleflux>
but yeah, it happens that I hit those BSD issues too :D
<mjg>
even if you only stick to linux
<particleflux>
it's bad enough that RHEL don't ships with term support for rxvt :/
<particleflux>
or rather, xvt-unicode-256color
<particleflux>
+r
<particleflux>
I don't even crae about the unicode or 256color things, I just want all my keys to work :D
invalidopcode has quit [Remote host closed the connection]
invalidopcode has joined #osdev
<particleflux>
hu, I've never seen an invalidopcodeexception
<kof123>
eh, im sure they tested gnome or kde or whatever other terminal and called it a day
dude12312414 has quit [Ping timeout: 255 seconds]
<kof123>
if it is just adding an entry that is kind of silly though
<kof123>
like what could it possibly break, something misidentifying itself?
nexalam has joined #osdev
nexalam has quit [Remote host closed the connection]