<mcrod>
is it possible that linux can just fucking work for fucking once
<mcrod>
older than I am and it still can't get fucking anything right that windows and apple solved before I was born
<mcrod>
jesus christ
<mcrod>
mac*
Burgundy has quit [Ping timeout: 246 seconds]
nyah has quit [Quit: leaving]
heat has joined #osdev
<heat>
hi its me warmth
<heat>
00:05 <mcrod> is it possible that linux can just fucking work for fucking once
<heat>
no
<mcrod>
boy I can't wait to go to #archlinux and ask why something as basic as audio doesn't work
<heat>
are you using arch?
<heat>
also ITS YER FAULT YOU DUMB DUMB
<Mutabah>
pulseaudio issues?
<mcrod>
this is what I did
<mcrod>
a) move to a different apartment
<Cindy>
mcrod: more like linux mint chat
<mcrod>
b) audio worked before this
<heat>
use pipewire
<heat>
pipewire works okayish for me
<Cindy>
linux mint is actually worse than arch linux in tech support
<mcrod>
c) not use front audio, but use line out in rear
<mcrod>
no audio
<zid>
I just use alsa cus you know.. it's alsa?
<zid>
sure I can't runtime configure stuff buut
<Cindy>
their distro is held together with duct-tape, any slight deviation causes it to fall apart and they'll tell you it's your fault
<zid>
who cares, set default sink in dmix and done forever
<Cindy>
what more do you expect from packages that haven't been updated in decades? :P
<heat>
you could be using windows
<heat>
audio workie? haha nice
<Cindy>
the problem with linux mint that i have, is outdated packages and i mean really REALLY outdated packages
<Cindy>
when you want to update something unofficially, you'll have to update the dependencies too, and dependecies of those dependencies
<Cindy>
and it snowballs from there
<zid>
audio on windows works unless you're on w10/w11 from what I've heard
<mcrod>
audio on windows worked straightaway, like I expected it to work
<zid>
windows update and nvidia updates etc just constantly fucking with people's audio setups
<zid>
cus you don't admin the machine, they do
<mcrod>
linux of course, as usual, doesn't work.
<mcrod>
no wonder companies don't embrace FOSS more
<mcrod>
most of it is *garbage*
<Cindy>
like jesus chrsit
<Cindy>
i just want to update latex to properly export to pdf
<Mutabah>
except valve, and 90% of internet hosting
<mcrod>
at least the proprietary *garbage* that is developed is generally well tested
<Mutabah>
windows, well tested?
<mcrod>
listen
<mcrod>
I worked in the semiconductor industry, where the software is being used to power bonding of large company X
<mcrod>
I can assure you
<mcrod>
it is some of the worst code I have ever seen written by the mind of man
<Cindy>
mcrod: code fast, bro
<mcrod>
I don't know how or why large company X is paying for it, but somehow, SOMEHOW it compiles, and works
<Cindy>
company doesn't care if the whole software is a pile of shit
<Cindy>
if it works, it works
<mcrod>
that's right
<mcrod>
that's what I'm saying
<mcrod>
it is horrifingly bad
<mcrod>
thank god I left that place
<mcrod>
but further what I'm saying is, people will pay real $$$$$$ for a pile of dog shit that works
<Cindy>
zid: i heard microsoft has already laid off most of their staff, replaced them with either out-sourced workers or "AI"
<Cindy>
they already laid off their OS quality control department
<mcrod>
it is 2023, linux distributions still can't figure out audio when (even though I'm sure the implementations are horrible in windows), windows has
<Cindy>
i remember this month when microsoft news pushed out an article made by a large language model
<Cindy>
where it recommended a food bank if you are hungry
<zid>
had broken and shitty audio its entire lifetime
<zid>
that's why people use *macs* for audio
<zid>
not windows
<mcrod>
I plugged shit into my ports and it worked automatically
<mcrod>
I did the same on Linux and it did not
<mcrod>
shitty or not in the details, one worked, the other didn't
<mcrod>
and ultimately that's what it comes down to
<mcrod>
if you're the type of guy who likes to fiddle with 50000000 config files to fix what is given by birthright from systems 20 years ago, fine, do you
<zid>
windows is *exactly the same*, a random subset of feautres will automatically work
<zid>
a random subset won't but on windows it will be impossible to fix
<mcrod>
random subset that just happens to work for 98% of use cases, considering most of the world doesn't seem to bitch about it
<kazinsal>
ironically the audio stack on openbsd seems to work a lot better than the one on linux, but I'm sure rewriting the entire linux audio stack a tenth time will make it work this time
<zid>
I hear *constant* bitching about audio on windows
<mcrod>
the remaining 2% is what to your point, people go out and buy macs for
<mcrod>
overall what I'm getting at is: shit code that solves the problem is better than good code that doesn't
<zid>
I searched the last discord I had open for windows audio
<heat>
(translation: lots of FOSS is complete ass, including most desktop software)
<heat>
but the problem truly is BKAC because you're fucking voluntarily using it anyway
<mcrod>
shjuo/ws 12
<mcrod>
uh
<mcrod>
no one saw that.
<kof123>
i'm still waiting for the riveting pci option rom discussion. how large may these be? how many maximum can be used? and so forth
<zid>
[02:20] -NickServ- Invalid password for mcrod.
<mcrod>
zid: :)
<mcrod>
that wasn't the password, but
<mcrod>
nice try
morgan has quit [Read error: Connection reset by peer]
morgan has joined #osdev
[itchyjunk] has quit [Read error: Connection reset by peer]
nvmd has quit [Quit: WeeChat 3.8]
<heat>
kof123, they may be as large as your device's flash
<heat>
but AFAIK they have the practical limit of 4GB (32-bits?)
k0valski18891 has quit [Ping timeout: 248 seconds]
<heat>
and you can have as many as you'd like, as long as they fit
morgan has quit [Read error: Connection reset by peer]
<zid>
That's what I tell the girls about sharpies
<zid>
(have you seen that subreddit? :P)
morgan has joined #osdev
<heat>
no
<zid>
you should check it out, 10/10
<heat>
what's the name?
<zid>
google for reddit sharpies subreddit
<heat>
haha that's brilliant
<heat>
truly peak humanity
<immibis>
[02:46] -NickServ- You are now identified as zid.
<zid>
my password is awful that doesn't surprise me
<immibis>
why is your first assumption that i know your password and not that i just typed something that didn't happen
<zid>
s/doesn't/wouldn't
<Mutabah>
Don't google that at work.
<heat>
is your password buttsharpies123
<zid>
123? that's a lot!
<immibis>
[02:48] -NickServ- You are now identified as Mutabah.
<immibis>
er
<immibis>
[02:48] -NickServ- You are now identified as heat.
<immibis>
damn i messed that up
morgan has quit [Read error: Connection reset by peer]
morgan has joined #osdev
morgan has quit [Ping timeout: 246 seconds]
morgan has joined #osdev
heat has quit [Ping timeout: 252 seconds]
Hammdist has joined #osdev
Affliction has quit [Quit: Read error: Connection reset by beer]
Affliction has joined #osdev
orccoin has joined #osdev
bliminse has quit [Ping timeout: 245 seconds]
bliminse has joined #osdev
torresjrjr__ has joined #osdev
corecode_ has joined #osdev
deflated8837_ has joined #osdev
Benjojo_ has joined #osdev
ggherdov_ has joined #osdev
leon_ has joined #osdev
Stary_ has joined #osdev
dennisschagt_ has joined #osdev
samis has joined #osdev
ggherdov has quit [*.net *.split]
torresjrjr_ has quit [*.net *.split]
dennisschagt has quit [*.net *.split]
jeaye has quit [*.net *.split]
deflated8837 has quit [*.net *.split]
CompanionCube has quit [*.net *.split]
Benjojo has quit [*.net *.split]
Stary has quit [*.net *.split]
leon has quit [*.net *.split]
corecode has quit [*.net *.split]
deflated8837_ is now known as deflated8837
Benjojo_ is now known as Benjojo
ggherdov_ is now known as ggherdov
jeaye has joined #osdev
jjuran has quit [Ping timeout: 252 seconds]
Hammdist has quit [Quit: Client closed]
bgs has joined #osdev
Hammdist has joined #osdev
GeDaMo has joined #osdev
Valeria22 has quit [Quit: Konversation terminated!]
frkzoid has quit [Ping timeout: 246 seconds]
jjuran has joined #osdev
m3a has quit [Ping timeout: 245 seconds]
stolen has joined #osdev
goliath has joined #osdev
benlyn has joined #osdev
bgs has quit [Remote host closed the connection]
dude12312414 has joined #osdev
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
d3ph4u17 has joined #osdev
corecode_ is now known as corecode
d3ph4u17 has quit [Client Quit]
Left_Turn has joined #osdev
SGautam has joined #osdev
Turn_Left has joined #osdev
Left_Turn has quit [Ping timeout: 246 seconds]
danilogondolfo has joined #osdev
stolen has quit [Quit: Connection closed for inactivity]
zxrom has quit [Ping timeout: 252 seconds]
nyah has joined #osdev
gog has joined #osdev
zxrom has joined #osdev
Burgundy has joined #osdev
benlyn has quit [Remote host closed the connection]
<Ermine>
gog: may I pet you
<gog>
Ermine: yes
* Ermine
pets gog
* gog
prr
<Ermine>
No heat here :(
<gog>
:(
zxrom has quit [Quit: Leaving]
zxrom has joined #osdev
gildasio has quit [Remote host closed the connection]
gildasio has joined #osdev
stolen has joined #osdev
orccoin has quit [Ping timeout: 252 seconds]
Hammdist has quit [Quit: Client closed]
<sham1>
Brrr
vdamewood has joined #osdev
dennis95 has joined #osdev
<mcrod>
gog may I hug you
Hammdist has joined #osdev
Left_Turn has joined #osdev
Turn_Left has quit [Ping timeout: 248 seconds]
Turn_Left has joined #osdev
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
Left_Turn has quit [Ping timeout: 260 seconds]
<Hammdist>
I have screen open on the serial port /dev/ttyUSB0. how would I write a file to the serial port as if I had typed it into screen?
<gog>
mcrod: yes
<gog>
Hammdist: cat file > /dev/ttyUSB0
<gog>
(meow)
MRtecno98 has joined #osdev
<Hammdist>
I had tried that first, but it had seemed not to work. but now I realize I haven't hooked up serial port receive in the program yet! so it probably works to just cat to the serial port
<gog>
everything is a file except things that aren't files
<Hammdist>
ceci n'est pas un fichier
<gog>
also idk what your device permissions config is like, if you need to change the perms on it or not
<gog>
or add yourself to a group
<zid>
I am not a french file either
Hammdist has quit [Quit: Client closed]
stolen has quit [Quit: Connection closed for inactivity]
<sham1>
cat /home/gog
<mcrod>
hi
* mcrod
hugs gog
* gog
prr
<mcrod>
my cat sneezes a lot
<sham1>
F
Hammdist has joined #osdev
<Hammdist>
so I have an ascii encoded write operation format that I use to send a program over the serial port. only about 98% of such writes are actually received. bummer
SGautam has quit [Quit: Connection closed for inactivity]
CryptoDavid has joined #osdev
orccoin has joined #osdev
Burgundy has quit [Remote host closed the connection]
Nixkernal has joined #osdev
gog has quit [Remote host closed the connection]
goliath has quit [Quit: SIGSEGV]
gog has joined #osdev
bgs has joined #osdev
<gog>
something is pretty fucked witho our internet
Valeria22 has joined #osdev
<zid>
Well it can't be a tree having fallen onto a phone line
<zid>
so I'm stumped
freakazoid332 has joined #osdev
<mcrod>
gog: the fishies ate the undersea cables
<Hammdist>
I found the problem with my serial port protocol: there was a command that would clear a memory region. I hadn't realized that takes a lot of time so it's necessary to stall the protocol with dummy data while it executes. now things are more or less working
<gog>
yay!
<gog>
mcrod: maybe i should eat the fishies
<gog>
but that would mean going into the ocean
<gog>
yuck
<Ermine>
gog paradox
<zid>
why not just use flow control?
<gog>
i'm only ON/OFF
<gog>
zid: isavia is trying to make the city cut down a bunch of trees in öskuhlið
<gog>
nobody is happy about this since we have no other actual wooded areas
<zid>
isavia?
<Ermine>
how that's spelled
<gog>
state airport operator
<zid>
ah
<zid>
island aviation
<zid>
makes sense now I know
<gog>
they want to get a little more shallow glideslope but the problem isn't the trees, it's that there's a hill that's 168' AMSL like a mile before the threshold
<zid>
They should cut a giant slice out of it, and leave the trees alone
<gog>
yes
<zid>
lower them back down on jacks
<Ermine>
Are there protests
<gog>
not really
<gog>
it's only a proposal rn
<gog>
the city said no
<zid>
I also said no
phoooo has joined #osdev
<Ermine>
heat
<Ermine>
sorry, misclicked
MRtecno98 has quit [Quit: Leaving]
<mcrod>
...
<mcrod>
today I learned that you do not need to use -dump-config for clang-format
<mcrod>
and that's probably true of clang-tidy
<mcrod>
I'm mad
<mcrod>
one of the first commands the manual tells you to use involves -dump-config
<zid>
it'd save time for you to tell us when you're not mad at something
<mcrod>
i'm not mad at you, or gog
<zid>
don't call gog a something
<mcrod>
wut
<gog>
yeah, i'm a nothing
* mcrod
hug gog
* gog
hug mcrod
phoooo has quit [Quit: Client closed]
SGautam has joined #osdev
stolen has joined #osdev
Jari-- has joined #osdev
gog has quit [Quit: Konversation terminated!]
gog has joined #osdev
Jari-- has quit [Remote host closed the connection]
apokryptein has joined #osdev
<dzwdz>
any ideas why ld wouldn't recognize an .ar? /usr/lib/libc.a: file not recognized: file format not recognized
<dzwdz>
this is a crosscompiled ld running on my os, the same exact command works on the host
<zid>
It'll say that for wrong arch or bitness
<zid>
cus ld's error messages are A+
<GeDaMo>
What does file say about it?
<dzwdz>
GeDaMo: i don't have it crosscompiled, but on the host it says that it's an ar archive
<dzwdz>
i didn't hash it from inside the vm or anything, but the header looks correct in there too
<dzwdz>
zid: i don't think it's just an ld thing, objdump fails too, with the same error message
<zid>
they're both binutils
<dzwdz>
yeah
<zid>
you could always un-ar it
<zid>
and check it's full of .o files for the right arch etc
<dzwdz>
what exactly would i be looking for
<zid>
.o files for the right arch etc
<dzwdz>
ar: /usr/lib/libc.a: file format not recognized
<dzwdz>
hmm
<dzwdz>
i'll check if the archive isn't corrupted anywhere
<dzwdz>
if it's possible that something went wrong when compiling it
<nortti>
do all of these use bfd?
<zid>
he doesn't get far enough :P
<dzwdz>
nortti: i think that message is coming from bfd
<dzwdz>
i don't remember finding it anywhere else in the binutils tree but i might be misremembering
<zid>
ar don't give no fucks about bfd
<dzwdz>
yet "file format not recognized" is only present in bfd/bfd.c
gog has quit [Quit: Konversation terminated!]
Hammdist has quit [Ping timeout: 246 seconds]
<geist>
maybe it’s some sort of bug in your fs layer that shows up with particular access patterns by different tools
<geist>
might make sense to port a hash utility or something so you can get a first pass validation of a file
<dzwdz>
i did, the crc32 checks out
<dzwdz>
my current assumption is that something went terribly wrong during the binutils build
<dzwdz>
so i'm trying to get it to build properly
<zid>
but it's broken in ar too you said
<dzwdz>
so?
<dzwdz>
it's part of binutils too
<zid>
oh, didn't realize ar came from binutils, til
<zid>
I'd just grab the file off and make sure a different ar can read it
gog has joined #osdev
<zid>
then depending on answer it's broken file or broken ar, then you test 'broken' ar with a known good file, etc
<dzwdz>
the file is known good
<dzwdz>
i mean the libc
<dzwdz>
the crc32 matches the version on my host, and it was used to compile several things
<zid>
so it opens fine in a different copy of ar?
<dzwdz>
yup
<zid>
ar broken then yea
<dzwdz>
i also just confirmed it on a clean build
<dzwdz>
so i'm guessing that it does some weird io that breaks my libc and/or kernel
vdamewood has joined #osdev
<zid>
ooh yea it doing weird fileops would defo not help
Yoofie5 has joined #osdev
<zid>
like if your libc isn't posixy enough
Yoofie has quit [Ping timeout: 260 seconds]
Yoofie5 is now known as Yoofie
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
dennis95 has quit [Quit: Leaving]
CryptoDavid has quit [Quit: Connection closed for inactivity]
<dzwdz>
i tried logging the file operations and dumping the output of each fread
<dzwdz>
but that looks correct too
<zid>
Time to instrument ar? :P
<dzwdz>
it looks as if it's trying to match the first file header against something but failing
<dzwdz>
as in the header of the first file stored in the archive
<dzwdz>
er
<dzwdz>
the first file header in the archive
<gog>
meow
* dzwdz
makes a human noise
<zid>
mmm kebab, definitely needed chilli sauce but there's no way I'm risking that rn
orccoin has quit [Ping timeout: 245 seconds]
<dzwdz>
bfd seems to be using something called "_bfd_check_format"
<dzwdz>
grepping for that across the binutils repo, i only find it being used
<dzwdz>
but it isn't defined anywhere
<dzwdz>
in other words i'm having a great time with this
SGautam has quit [Quit: Connection closed for inactivity]
<sham1>
*human noise*
gog has quit [Ping timeout: 240 seconds]
stolen has quit [Quit: Connection closed for inactivity]
PapaFrog has quit [Ping timeout: 256 seconds]
GeDaMo has quit [Quit: That's it, you people have stood in my way long enough! I'm going to clown college!]
apokryptein has quit [Quit: Leaving]
<dzwdz>
any tips except instrumenting ar?
gorgonical has joined #osdev
<dzwdz>
the control flow in bfd is horribly obfuscated, i can't bring myself to deal with it any more
goliath has joined #osdev
<nortti>
if you have a debugger running on your own OS it might be useful comparing the control flow between your host system and your OS and seeing where they diverge
<dzwdz>
i don't
<dzwdz>
i can attach to qemu with gdb, but that breaks on context switches and so far i wasn't able to figure out how to solve that
<mcrod>
hi
<nikolar>
hello
Nixkernal has quit [Ping timeout: 260 seconds]
<gorgonical>
Any arm experts mind helping me understand secure vs non-secure access to the gic?
<gorgonical>
I'm booting a secure kernel on core 1 and it brings the gic up enough to have timer interrupts, and then on core 0 I let u-boot run enough to wake up Linux. During bootup Linux also configures the gicd because it thinks its the first kernel on the system. But it reconfigures the gicd_ctlr register and turns *off* G1S interrupts.
<gorgonical>
That's expected, but I don't even know why Linux can see that version of the gicd_ctlr. The manual suggests that non-secure access should see that bit (G1S enable) as res0. So then if Linux writes 0 to gicd_ctlr while it reconfigures, shouldn't that write do nothing?
<gorgonical>
Alternatively, am I misunderstanding what counts as "secure access"? I am 95% sure Linux is not being run with SCR_EL3 configured so Linux is secure, so is there some bus configuration I'm missing? Anyone with more experience/insight?
<clever>
and its up to EL3 to decide the NS and EL it eret's into
<clever>
the ideal being, your secure kernel and non-secure linux can co-exist on core0
<clever>
a GIC timer interrupt could get routed to EL3, which will context switch would then context switch between the 2 kernels
alexander has quit [Quit: ZNC 1.8.2+deb3.1 - https://znc.in]
<clever>
or it could be entirely event driven, linux issues an SVC that traps into EL3, which eret's down into the secure EL1 to do something
<clever>
SMC*
alexander has joined #osdev
<gorgonical>
clever: i know about the same-core segregation, but we want a full separation for availability reasons
<clever>
youll still need to ensure core0 is in NS mode, while core1 is in S mode
<gorgonical>
my understanding is that the timer exists at each core and is in fact a ppi, so afaict it shouldn't even be affected by the gicd settings
<gorgonical>
yeah and I think we're doing that. we have u-boot as the main bootloader with arm tf-a to run the secure payload. I haven't 1000000% verified it but my understanding is that u-boot will drop linux into ns el1. I know it's el1 at this point
<gorgonical>
but is it your understanding that "secure access" is reliant on the PE's NS/S bit?
<clever>
gorgonical: its also reliant on the chip OEM to wire the NS flag from the arm core, up to other peripherals
<clever>
some chips just dont have a working secure mode
<gorgonical>
I am hoping that's not the case. The chip's TRM *says* it has a working secure mode
<gorgonical>
My suspicion when I got up this morning was that I needed to configure the bus that the GIC is attached on. The chip says the gic is across the APB bus, so maybe there's some configuration there?
<gorgonical>
But I don't know if that's actually well-founded
<gorgonical>
the manual says, not the chip
gog has joined #osdev
PapaFrog has joined #osdev
Left_Turn has joined #osdev
Turn_Left has quit [Ping timeout: 248 seconds]
zxrom_ has joined #osdev
zxrom has quit [Read error: Connection reset by peer]
[itchyjunk] has joined #osdev
<dzwdz>
i just want to share that i figured out the binutils issues
<dzwdz>
and managed to actually compile something on my os
<nortti>
nice
<nortti>
what was the issue?
<nortti>
also, do you have the source code somewhere where I could test it out?
<dzwdz>
i'm kinda ashamed to admit it was a stubbed out sscanf
<dzwdz>
that i forgot to make throw an error
<dzwdz>
i do, but right now it doesn't include the binutils port yet
<nortti>
< dzwdz> i'm kinda ashamed to admit it was a stubbed out sscanf ← is that why you ended up looking at the float foramtting?
<nortti>
*formatting
<dzwdz>
sort of - i grabbed sortix's scanf, which made me wonder about how you guys are handling printf
<dzwdz>
so i looked at that, noticed that %f wasn't implemented, and remembered that i've wanted to try implementing float formatting for a whlie
Thedarkb has quit [Ping timeout: 258 seconds]
<dzwdz>
so how big of a pain is crosscompiling gcc
<dzwdz>
binutils was quite the experience already
heat has joined #osdev
<heat>
-SaslServ- Last failed attempt from: immibis!~quassel@2a01:4f9:4a:4caf::2 on Aug 24 01:48:48 2023 +0000.
<heat>
bruh
<gorgonical>
lol wat
<immibis>
oh no, not my ip address
<immibis>
you can ddos me with that
<heat>
im going 2 hack u
<dzwdz>
imagine having ipv6
<immibis>
i don't have ipv6... my quassel server has ipv6, there's a difference
<gorgonical>
I have lost so much hair this week from this dumb gic problem
<samis>
dzwdz: imagine *not* having ipv6
<immibis>
now I wonder about leaving secret notes to people by failed login attempts
<dzwdz>
i think technically speaking my isp does support ipv6
<dzwdz>
but you get either ipv4 OR ipv6
<nortti>
what
<dzwdz>
at least according to what i've read online
<immibis>
people in developed countries with wired connections don't have ipv6 because ISPs are averse to nonbroken internets
<dzwdz>
nortti: yeah, idk either
<immibis>
my connection is not wired, so I have one at home
<samis>
what, no DSLite or equivalent?
<dzwdz>
but it's not like i actually even bothered to ask them, and people on the internet do have a tendency to say dumb stuff sometimes
<heat>
i have v4 and v6
<heat>
succit
<zid>
I have v4 and a v6 tunnel :P
<immibis>
hurricane electric?
<zid>
he hype
<samis>
(mildly interesting: providing ipv6, but not an ipv6 dns server, totally makes sense. Don't really use the ISP DNS anyway
<zid>
german isps are all dslite from what I've seen, which is annoying when we wanna gaming
<dzwdz>
i think my (isp-supplied) router is too shitty to use the he tunnel
<zid>
use it on your machine then
<dzwdz>
doesn't it use some funky protocol
<dzwdz>
that i can't just port forward
<zid>
yes, ipv4
<clever>
after my last modem/router upgrade, i wound up with an SFP module slotted directly into my ISP router
<samis>
what kind of isp router takes an sfp module
<clever>
i would need an SFP adapter board to replace the isp router
<clever>
samis: one offering 1500mbit of download, lol
<dzwdz>
zid: 6in4, to be exact
<dzwdz>
which i believe my router needs to be able to understand
<zid>
proto 42 instead of proto 40 or whatever
<dzwdz>
otherwise it'll just drop the packets
<zid>
your router needs to not actively drop it yes
<zid>
That is true of all parts of tcp
<zid>
udp, dns, icmp, etc
gog has quit [Quit: byee]
gorgonical has quit [Ping timeout: 245 seconds]
<dzwdz>
my router only allows me to forward tcp and udp so i've kinda assumed that wouldn't work
mhall has quit [Quit: Connection closed for inactivity]
<dzwdz>
but actually lemme try
<immibis>
german ISPs allocate public IPv4s to customers because it is a developed country which means they got allocated IPv4s before exhaustion was a thing
<zid>
works fine on everything I've seen
<zid>
immibis: incorrect
<immibis>
are you using an old ISP or a new startup one?
<zid>
You're thinking of the UK, germany is all dslitey
<dzwdz>
FAQ: I'm trying to set up a tunnel to an IP behind a router providing NAT. Will this work? / In most cases, this configuration will not work. [...]
<zid>
100% lies
<zid>
it will almost always work
<zid>
like, the only place I remember that mentioned when I was setting it up was
<zid>
NOTE: When behind a firewall appliance that passes protocol 41, use the IPv4 address you get from your appliance's DHCP service instead of the IPv4 endpoint you provided to our broker.
<zid>
"make sure you use your nat address bro"
<immibis>
IP over IP? That won't be port forwarded without customized router firmware
<immibis>
only UDP and TCP get NATted
<immibis>
DMZ might work if your router has it
<zid>
good news, it doesn't need it
<immibis>
IP over UDP would work
<dzwdz>
wireguard ftw
Left_Turn has quit [Read error: Connection reset by peer]