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
<mcrod> gog may I pet you
<gog> yes
nyah has quit [Quit: leaving]
* mcrod pets gog
* gog prr
melonai has joined #osdev
mhall has quit [Quit: Connection closed for inactivity]
heat has quit [Remote host closed the connection]
heat has joined #osdev
<heat> this time i rebooted cuz i did <up arrow> + <enter> and the last command was reboot
<heat> oopsie
<mcrod> this vaguely reminds me when you bricked your EFI
<mcrod> or whatever that was
<heat> how
<heat> the situations are completely different
smokealot has quit [Quit: WeeChat 3.8]
<heat> anyway im writing unit tests
<heat> is this poggers
<heat> im adding augmentation to trusty's lk fork's lib/binary_search_tree
<heat> cuz i really need it for vm
duderonomy has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
vdamewood has quit [Read error: Connection reset by peer]
vdamewood has joined #osdev
<heat> geist: you probably want to pick up https://android-review.googlesource.com/c/trusty/lk/common/+/1126069/11 if you ever get to importing lib/binary_search_tree
<bslsk05> ​android-review.googlesource.com <no title>
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
mavhq has quit [Read error: Connection reset by peer]
MiningMarsh has quit [Ping timeout: 240 seconds]
mavhq has joined #osdev
roan has joined #osdev
diamondbond has joined #osdev
diamondbond has quit [Remote host closed the connection]
<geist> yeah been meaning to go back over that for some time
<heat> their solution still isn't great for allocation as they iterate the whole address space
<heat> so you kind of have O(n) allocation
gog has quit [Ping timeout: 246 seconds]
terrorjack has quit [Quit: The Lounge - https://thelounge.chat]
terrorjack has joined #osdev
dude12312414 has joined #osdev
dude12312414 has quit [Remote host closed the connection]
gorgonical has quit [Ping timeout: 272 seconds]
Matt|home has joined #osdev
duderonomy has joined #osdev
[itchyjunk] has quit [Read error: Connection reset by peer]
heat_ has joined #osdev
heat has quit [Read error: Connection reset by peer]
eddof13 has joined #osdev
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
_whitelogger has joined #osdev
kof123 has joined #osdev
stefanct has quit [Ping timeout: 252 seconds]
stefanct__ has joined #osdev
stefanct__ is now known as stefanct
heat_ has quit [Ping timeout: 264 seconds]
slidercrank has joined #osdev
Gooberpatrol66 has joined #osdev
bradd has quit [Quit: No Ping reply in 180 seconds.]
bradd has joined #osdev
meisaka has quit [Ping timeout: 260 seconds]
meisaka has joined #osdev
zxrom__ has joined #osdev
zxrom_ has quit [Ping timeout: 260 seconds]
slidercrank has quit [Ping timeout: 260 seconds]
zxrom__ is now known as zxrom
<vai> really neat idea for guys to make a Christmas special for the OS
bradd has quit [Ping timeout: 245 seconds]
goliath has joined #osdev
Gooberpatrol66 has quit [Ping timeout: 245 seconds]
Gooberpatrol66 has joined #osdev
qubasa has joined #osdev
vai has quit [Ping timeout: 264 seconds]
slidercrank has joined #osdev
bradd has joined #osdev
bradd has quit [Client Quit]
bradd has joined #osdev
Burgundy has joined #osdev
ZipCPU_ has joined #osdev
ZipCPU has quit [Ping timeout: 245 seconds]
ZipCPU_ is now known as ZipCPU
danilogondolfo has joined #osdev
GeDaMo has joined #osdev
nyah has joined #osdev
vai has joined #osdev
cloudowind has quit [Ping timeout: 272 seconds]
mhall has joined #osdev
<ddevault> now we're getting somewhere
<zid> what's the tippex writing say.. 1280x800x32 vesa?
<ddevault> details on the display, for use with 9front
<zid> what's the left hand side say
<ddevault> vgasize=1280x800x32 vesa
<zid> ah vgasize
<zid> I got vsa????
<ddevault> my handwriting has won no awards
floyd has joined #osdev
<Mutabah> Doesn't help that the writing is too small for the pen tip
<zid> and it's out of focus
<ddevault> ur out of focus
<zid> I've got too much month and not enough focii
<Mutabah> Not really out of focus, not compared to the keyboard.
<zid> the screen is out of focus too, and it's at the same depth as it
<zid> the camera seems to have picked about 4 inches in front of it
<zid> which makes the keyboard keys the most readable
<Mutabah> Focus isn't the best... but the writing is just too thick a marker for the size
<Mutabah> ddevault: Sorry if you don't like the discussion :)
* ddevault shrugs
<zid> I want one of those pens you write "£300 obo" onto car windscreens with
<zid> grease pen of some type
<zid> apparently, chalk
gog has joined #osdev
slidercrank has quit [Remote host closed the connection]
slidercrank has joined #osdev
Left_Turn has joined #osdev
<ddevault> excuse the shaky handheld cam https://l.sr.ht/eNqI.webm
awita has joined #osdev
<gog> nice!
<gog> i should get a beat up old computer to test on but i actually have one
<ddevault> have consistently tested on real hardware, want this to actually work
<ddevault> though until I finish x86_64 UEFI support I can only test on a small number of machines
zxrom has quit [Quit: Leaving]
<nortti> how is it booting there? MBR?
<ddevault> MBR
<ddevault> better video of qemu https://l.sr.ht/d_tH.webm
osmten has joined #osdev
kof123 has left #osdev [#osdev]
MiningMarsh has joined #osdev
rwb has quit [Ping timeout: 260 seconds]
rwb has joined #osdev
osmten has quit [Quit: Client closed]
smokealot has joined #osdev
rwb has quit [Ping timeout: 246 seconds]
heat_ has joined #osdev
rwb has joined #osdev
smokealot has quit [Quit: WeeChat 3.8]
awita has quit [Remote host closed the connection]
SGautam has joined #osdev
[itchyjunk] has joined #osdev
awita has joined #osdev
awita has quit [Client Quit]
gareppa has joined #osdev
<gog> hi
<sham1> hi
<heat_> hi
<Ermine> hi gog, may I pet you
<gog> yes
<mjg> no fraternizing during office hours
<Ermine> I'm on vacation
* Ermine pets gog
* heat_ pets mjg
<mjg> OH
* mjg goes to HR
<sham1> Unsolicited pets!
<heat_> OH
<Ermine> OH, heat_ soon will have more time for onyx!
<heat_> i already have all the time for onyx
<sham1> I wish I had time for programming
<sham1> …or motivation
<Ermine> I have enough time during holidays, but by that time I get exhausted enough to not have any motivation
<heat_> hi ermine where's your linux bootloader
<heat_> and where's my linux boot stub implementation?
<heat_> cheers mate
<Ermine> Maybe I should return to that
* gog prr
<gog> hi
<gog> i'm gonna program when i get home
<gog> i'm also programming right now
<gog> i program for money and for fun
<gog> i'm an anxious mess and my life has no meaning beyond code
<heat_> grog program
<gog> grogram
<heat_> grogramming
<gog> meow
<heat_> gog you're wanted and loved
<heat_> and you can accomplish things
gareppa has quit [Quit: WeeChat 3.8]
<gog> thanks <3
<gog> i had a frustrating bug i couldn't solve last night but the answer came to me in a flash of inspiration while i was on the toilet
<gog> it reminded me that i'm not totally incompetent
<mjg> is the solution.. shit?
<gog> no
<gog> but the solution is facilitated by shit
<sham1> It's always the toilet, the shower or being asleep that gives answers
<mjg> if in doubt, add zero-cost abstraction
<gog> so when you're creatively constipated
<gog> mjg: that's pessimal
<gog> abstractions are pessimal
<sham1> mjg: I thought you didn't like rust
<mjg> but it is ZERO COST!!
<mjg> next step is negative cost abstractions innit
<gog> optimiziations are negaative cost abstractions
<gog> or are optimizations actually concretions
<mjg> how to optimize, geezer guide
<mjg> 1. don't do shit, claim software is fast enough
<mjg> 2. if pressed, wildly speculate what the slowdown is and patch that
<mjg> 3. DO NOT OBTAIN PROFILING DATA
<mjg> 4. watch a young person do the actual work later
<mjg> go to 1
<mjg> feel free to claim greatness at any point
<mjg> if called out, claim pragmatic tradeoffs
<heat_> damn young people
<mjg> basically anything but admitting the code is shite
<sham1> I see you have also worked in acadsmka
<sham1> Academia*
<mjg> no, but i had displeasure working with teaching assistants going into industry 'n shit
<sham1> Uuh
<mjg> the fact this people /teach/, when they should have never graduated, sealed my opinion on degrees as a predictor of quality
<sham1> I work in research and I'm annoyed that I can't exactly disagree. Degrees are a bad proxy metric for software stuff
<mjg> i know for a fact that in poland, where higher education is free (as in taxpayer subsidized) it became a business to graduate people
Burgundy has quit [Ping timeout: 246 seconds]
<mjg> and i strongly suspect something of that sort is present in other countries
<mjg> at the end of the day you money going in is dependent on your student head count
<mjg> and if an idiot drops out, he may enroll elsewhere so you are giving money away
<mjg> to the competition
<sham1> Finland also does that. Universities get money for every student that graduates, and most people graduate to become MSc or equivalent which in turn creates degree infla5
<mjg> there are private "universities" in poland where you have pay tuition and which are de facto diploma millas
<sham1> Degree inflation
<mjg> mills
slidercrank has quit [Quit: Why not ask me about Sevastopol's safety protocols?]
<mjg> except takes years to get one
<mjg> you need to attend classes 'n shit, but you don't need to learn anything or do almost any work
<mjg> here is an example: there will be an exam you can take 4 times. first 3 times you get the same questions
<mjg> idiots who still did not pass get the 4 exam, where they show up to get the passing grade automatically
<mjg> this is literally the model, no joke
<mjg> during the 3 times where you have to write the exam you often allowed to cheat
<sham1> Okay yeah, that's not a proper academic institution by any metric
<mjg> it basically takes willingness to NOT pass to not pass
<mjg> i'm afraid this is quite widespread in poland
<mjg> not that /every single uni/ is like that
<mjg> but the count is high enough that ultimately if you want a degree, you can easily find a uni where you will get one by "passing" as described above
<mjg> rendering mere fact that someone has the degree worthless
<mjg> you can try to filter by uni, but people intentionally misrepresent what happened
<mjg> example: there was a uni with several different departments, offering 4 separate "applied computer science" degrees
<mjg> one of them is quit renonwed, hard to get in and demanding
<mjg> the rest is as described above
<mjg> so people who graduate from the 3 crappers skip mentioning the department
<mjg> :]
<SGautam> This is going to sound a bit of a shameless request, but have anyone of you written a VT100 emulator for the VGA 16 color backend
<mjg> implying they passed the demanding one
<heat_> SGautam, i have a vt100 "emulator", but not for vga
<heat_> rather just a normal framebuffer
<SGautam> I'm guessing at one point it must be outputting a pure text stream to a backend text buffer.
Brnocrist has joined #osdev
<heat_> yep
<heat_> see kernel/kernel/tty/vt/vterm.cpp in onyx
eddof13 has joined #osdev
<SGautam> thanks
eddof13 has quit [Client Quit]
goliath has quit [Quit: SIGSEGV]
vdamewood has joined #osdev
eddof13 has joined #osdev
<mcrod> test
<mcrod> hi
<gog> test
<mcrod> oh thank god
<mcrod> i am back on IRC from iOS
<mcrod> i think it’s working properly now anyway
<mcrod> gog may I pet you
<heat_> <mcrod> i am back on IRC from iOS
<heat_> cursed
<heat_> ping me on imessage ok mcrod?
<gog> mcrod: yes
<gog> mcrod: do u have discord
* mcrod pets gog
<mcrod> yes
<mcrod> mcroddev
<vai> klange: I am adapting your ELF32 dynamic linker to a Turbo C environment (protected mode)
<heat_> top 10 things that should never have been said
<mcrod> i wanna go home
<gog> i made progress on things today
<gog> i am a competent developer
<gog> i can be a messy bitch and a competent developer at the same time
<vai> klange: this must be it (all dependencies besides the system headers) :: #include "../lib/list.c" & #include "../lib/hashmap.c"
<bslsk05> ​github.com: toaruos/linker/linker.c at master · klange/toaruos · GitHub
<mcrod> i’m waiting for approvals from a team that we don’t have control over.
<bslsk05> ​reviews.freebsd.org: ⚙ D41092 libthr: Use a small cache to greatly reduce pshared lock latency.
<mjg> great quality all around innit
Gooberpatrol66 has quit [Ping timeout: 260 seconds]
dude12312414 has joined #osdev
dude12312414 has quit [Client Quit]
gog has quit [Quit: Konversation terminated!]
<heat_> i would really like for an easy way to get the origin IP when a jmp/call crashes
<Cindy> is there like one of those exception stack frames?
<sham1> I'd like to be able to easily access IP in general
<heat_> exceptions just tell you "you crashed trying to execute 0x0"
<heat_> which, ok, cool, but it doesn't tell me shit
<heat_> i believe all architectures i've worked with do this
<Cindy> doesn't the stack frame have registers in it?
<Cindy> like the instruction pointer
<heat_> sure, the new registers
<Cindy> and stack pointer and shit
<sham1> I get that there are reasons why ISAs don't always like to expose the IP registers, but it'd be nice
<heat_> jmp doesn't leave shit in the stack :/
<heat_> instead of saying IP=0 fault_address=0, it should say IP=originIP, fault_address = 0
<heat_> because if you get a execute page fault on fault_address = 0 you kind of get what happened but didn't lose part of the previous context
<Cindy> in m68k, i can get the origin IP from the stack frame
<sham1> The most you'll get on x86 and AMD64 is any callers
<Cindy> downgrade
<Cindy> heat_: jmp? you mean if you jump to invalid address?
zxrom has joined #osdev
<Cindy> or if the code you're jumping to causes a processor exception
<heat_> first
<Cindy> idk, but the first in m68k would cause either an address error or bus error
<heat_> unfortunately not many people are on m68ks these days
<sham1> I wish we had a 64-bit m68k
<sham1> Such a good processor
<Cindy> sham1: well, there IS an unused operand for the arithmetic and memory operations
<Cindy> you could use that for 64-bit
<Cindy> i mean implementing it
<Cindy> the size operand is 2 bits, but only 00, 01 and 10 are used
<sham1> Yeah, so 11 and it could be 64-bit
<sham1> That'd be nice
<sham1> Then we'd just have to find us a chip fab. Oh and get Motorola to not sue us to oblivion
<sham1> I'd rather get sued to Skyrim /s
<heat_> hehehe
<heat_> hehehehehe
<heat_> hehe
<heat_> he
xenos1984 has quit [Ping timeout: 240 seconds]
xenos1984 has joined #osdev
<sham1> It's the only other Elder Scrolls game I know
<heat_> ok so
<heat_> im confused bois
<heat_> but i did build a cool stack dumper in the meanwhile
<bslsk05> ​i.imgur.com <no title>
<heat_> i've been doing shit like this manually for SEVEN FUCKING YEARS
<sham1> ouch
<mjg> libcarpaltunnelsyndrome
<mjg> name for the lib inspired by its need
<mjg> :X
<Cindy> sham1: i'm sure there's that extended version of m68k on a FPGA
<Cindy> for amiga though
<sham1> 64 bit Amiga would be glorious
wootehfoot has joined #osdev
<Cindy> yes
<Cindy> just extend the registers to 64-bit, add 64-bit size operand support... and you got yourself a 64-bit m68k
<Cindy> (anything beyond 64-bit would probably need new instructions)
<sham1> 64-bit address space
<Cindy> yea
<Cindy> which means more pins in the CPU packaging
<sham1> Could always do banks
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<sham1> So it'd be like the x32 ABI for AMD64 where you have 64-bit registers but 32-bit pointers
<Cindy> nah, no banks
<Cindy> the simplicity of m68k is that you just need a memory address
<sham1> So one would just extend the CPU packaging then for more address and data pins
<Cindy> yeah
<Cindy> or you could do socketed
<Cindy> like intel CPUs
<nortti> alternatively could keep 32-bit physical address space but have a bigger virtual one
xenos1984 has quit [Ping timeout: 246 seconds]
eddof13 has joined #osdev
gog has joined #osdev
Gooberpatrol66 has joined #osdev
terminalpusher has joined #osdev
<sham1> So PAE
<sham1> Hm
<nortti> reverse PAE
<sham1> True
xenos1984 has joined #osdev
<gog> pae bae
<heat_> pae is super useless
<heat_> actually, its only useful because of NX
<zid> pae kinda useful on amd64
<zid> given it's, you know, mandatory
<sham1> That doesn't count
<zid> yea not really, but technically true though
<zid> there's the weird pre-pae thing too, what was that called
<zid> pae20 or something strange, that only the 386 supported or something
slidercrank has joined #osdev
cross has quit [Remote host closed the connection]
Burgundy has joined #osdev
<mjg> pae was kind of practical back in the day
<mjg> but it was mostly enabling you to suffer as opposed to not being able to run to begin with
<sham1> ItaniuM!
<vai> nortti: bigger pages would solve the 32 bit mode limitations
<nortti> how do you mean?
<zid> itanium 3 soon sham1, are you hyped?
<sham1> Lies!
<vai> nortti: >4 Gb RAM
<nortti> how?
<zid> That's entirely independent of page size
<vai> zid: nortti: I heard of 4 Mb pages
<vai> thats supported
<zid> VA space is governed by the number of layers, physical space is governed by how wide each entry is (and possibly limited by hw)
<zid> page size doesn't enter into it
<vai> zid: you mean the addresses will be still 32 bit ?
<zid> page size governs fragmentation and tlb use
<zid> You could have 128bits of VA and 4 bits of physical memory, if you wanted
<SGautam> Is the scrolling handled by the terminal emulator or tty?
<zid> I have 48 bits of virtual and about 35 bits of physical, atm
<SGautam> Generally speaking
<zid> SGautam: a real terminal is just a line printer
<zid> with a print head and a carriage
<nortti> SGautam: terminal emulator. tty is just a pipe that you shove bytes into and get them out of plus some structs
<heat_> SGautam, emulator
<zid> go look at curiousmarc's teletypes if you want
<bslsk05> ​'Using a 1930 Teletype as a Linux Terminal' by CuriousMarc (00:16:33)
<heat_> today in me: we found a gcc bug in #armlinux, i found out gold is broken and was making my libc crash, i was accidentally mapping ASAN shadow over my brk
<SGautam> I'm having a little bit of trouble visualising how a program like vim works
<heat_> today has been a weirdo day
<SGautam> If the tty output is just a line
<zid> vim itself spits out a VT-100 command stream
<heat_> everything has been fucked but nothing was my problem
<zid> to make a physical vt-100 do the right thing
<zid> that is, move the cursor around, display characters, etc
<zid> that goes through the 'tty' layer, so that your terminal emulator can find the other end of the pipe, then the terminal emulator decodes the ansi commands
<vai> my Commodore 128 used to support move screen up (PETSCII) if I recall
<zid> and renders the text as though it were a scrolling ream of paper
<zid> There's a wrapper library for VT-100 commands too you may be familiar with, ncurses
<SGautam> Yes.
<zid> mvprintf(x, y, "blah"); -> printf("\e[%d;%dH%s", x, y, str);
<zid> terminal emulator just looks for \e and \r and \n and \t and stuff and processes them
<sham1>
<heat_> sham
<heat_> sham ar eyou ok?
<sham1> no
<heat_> ok
dude12312414 has joined #osdev
<heat_>
<zid>
<zid> I'd just like to take a moment to appreciate the TELNET protocol btw
<zid> I think it's quite nice
<mjg> :[
<gog> tell net
<heat_> no, dont tell her anything
<gog> i already did
<zid> lol, friend just told us about a fire he had to put out at work today
<zid> he works for a gambling website, some random fired up the stress testing tool they have, which creates loads of open bets and fills them and stuff as fast as it can
<zid> ..someone pointed it at the live DB
vdamewood has quit [Ping timeout: 246 seconds]
<zid> so the website was full of open, system created bets
<nortti> lol
<zid> > Congratulations to the one punter that figured it out fastest and made off with £8k.
goliath has joined #osdev
<gog> lmao
<gog> amazing
zxrom_ has joined #osdev
zxrom has quit [Ping timeout: 272 seconds]
terminalpusher has quit [Remote host closed the connection]
nvmd has joined #osdev
nvmd has quit [Client Quit]
nvmd has joined #osdev
<gog> hi
<mjg> zid: sounds like at least 2 fails there
<mjg> primary being prod db/api/whatever teh fuck reachable from test env
<mjg> which is also the classic
<zid> again?
<zid> we just had a hi last week
<zid> Such a demanding pet
* mjg throws gog a dead bird
* gog chomp bird
<mjg> no go kill me some pigeons
<mjg> you get to eat them
<gog> no thanks
<mjg> gg
<bslsk05> ​'Himari, Nom nom nom nom' by Fuzzychale Lingatong (00:00:15)
<heat_> mjg, openbsd!
vdamewood has joined #osdev
Gooberpatrol66 has quit [Ping timeout: 246 seconds]
<sham1> Is that the safeword?
<gog> hhhhh
<gog> nom nom nom nom
<heat_> sham1, oh yeah i'm gonna open the shit out of your bs-d
Gooberpatrol66 has joined #osdev
<gog> heat
<heat_> gog
<gog> hi
<heat_> hi
<heat_> hi gog, decode this: std::__2::vector<std::__2::basic_string<char, std::__2::char_traits<char>, std::__2::allocator<char> >, std::__2::allocator<std::__2::basic_string<char, std::__2::char_traits<char>, std::__2::allocator<char> > > >::~vector[abi:v15000]()
<heat_> thanks
<gog> that's just a vector of basic char strings
<bl4ckb0ne> i think you have c++
<gog> call me when you've got a vector of basic bitches
<bl4ckb0ne> its terminal
<gog> true
<gog> ther'es only one outcome for c++ programmers
<heat_> no, it's GUI
wootehfoot has quit [Quit: pillow time]
<heat_> hehehe he
<gog> it's always hehe and never sheshe smh
<heat_> ok gog i need you
<heat_> now its 4 real
<heat_> i have weird corruption i cant track down on my kernel test suite that *forks*
<heat_> i don't know if i should start an actual bisect
<gog> forking?
<heat_> doing a bisect would implicate me cleaning up my current local tree
<heat_> which has at least 3 or 4 unrelated change sets going on at the same time
<heat_> https://github.com/heatd/Onyx/pull/74 <-- it's all going to this PR which has... 37 commits
<bslsk05> ​github.com: Io queues by heatd · Pull Request #74 · heatd/Onyx · GitHub
<gog> yikes
<gog> stash everything and start cherry picking
<mjg> soudns like bisect would be faster
<bslsk05> ​i.imgur.com <no title>
<heat_> help me
<heat_> mjg, bisect needs a clean tree yeah?
<mjg> probably
<gog> oh my gog
<gog> have fun in git hell
<mjg> how did you fick this up so bad
<gog> i'm pretty bad but i never let it get like that
<mjg> as a single person working on the project
<gog> that's how
<gog> nobody to enforce proper git usage on him
<heat_> well
<mjg> smh my head
<heat_> i have many changes that kind of rely on each other
<mjg> then topo sort them and implement in that order
<heat_> like i built a whole tracing-ish thing that helped me take a look at some internals stuff
<gog> cherry pick and rebase
<mjg> use CVS
<gog> hhhhhhh
<gog> concurrent versions
<mjg> cvs branching == $$
GeDaMo has quit [Quit: That's it, you people have stood in my way long enough! I'm going to clown college!]
<zid> rebase baby
<zid> only tool you need
<heat_> ok its almost all done
<heat_> wasn't hard
<heat_> this PR has 45 commit of unrelated garbo
<heat_> masterpiece
<zid> see, rebase
<zid> don't p those ones
floyd is now known as cloudowind
<mjg> git commit -am "wip"
<mjg> no rebase
wblue has joined #osdev
<heat_> oh look i have a random copy of opendtrace i never got to integrate into onyx
<zid> heat traces a lot of d afterall
<sham1> Squash rebase PRs if you're a coward
<zid> unless you're a coward*
<mjg> it takes ballz to yolo the codebase
sortie has quit [Remote host closed the connection]
<sham1> Show your merges and commits!
Turn_Left has joined #osdev
<cloudowind> mmm
sortie has joined #osdev
Left_Turn has quit [Ping timeout: 260 seconds]
<cloudowind> freenode was hiding our ip's right? libera doesnt
<gog> show me your pull requess, your pull requests, show me your pull requests
<gog> cloudowind: you need a registered account and then you go to #libera-cloak and say !cloakme
<cloudowind> good practice to get a cloak tho
<gog> oh you have a cloak
<cloudowind> right
wereii has quit [Quit: ZNC - https://znc.in]
<gog> your IP is hidden to me
<cloudowind> yes yesterday i got one
<gog> good
<cloudowind> ty
<gog> np
wereii has joined #osdev
<mcrod> hi
<gog> hi
<cloudowind> hi
<mcrod> gog i want sleep.
<gog> so go slep
<cloudowind> he might have soe missing sheeps
sortie has quit [Ping timeout: 272 seconds]
<cloudowind> *some
heat has joined #osdev
heat_ has quit [Read error: Connection reset by peer]
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
<mcrod> i honestly think I have sleep apnea
<mcrod> apparently I snore something fierce and I wake up feeling tired all of the time
<gog> dang
Gooberpatrol66 has quit [Quit: Leaving]
<cloudowind> my gf got one
<cloudowind> she got better since she quit drinking and smoking tho
<cloudowind> otherwise she was waking up every hour struggling to breathe...
<cloudowind> you would know it if you had one.. getting up tired is not by itself a marker for it anyway you guys have a beautiful night
MiningMarsh has quit [Quit: ZNC 1.8.2 - https://znc.in]
MiningMarsh has joined #osdev
danilogondolfo has quit [Remote host closed the connection]
<heat> gog: here are my pull requests: https://github.com/heatd/Onyx/pulls
<bslsk05> ​github.com: Pull requests · heatd/Onyx · GitHub
<heat> io queues started as, erm, io queues, but ended up as some sort of staging branch for random crap
<heat> the arm64 PR is pretty tidy and related to arm64, but still large because it's a brand new arch
<heat> the ext4 stuff im basically just postponing endlessly
foudfou has quit [Ping timeout: 240 seconds]
<heat> the other PR was a bot
<heat> I hope you liked a tour through my PRs
foudfou has joined #osdev
<moon-child> ohh did you implement io_uring?
<heat> no
<heat> far more basic
<moon-child> :\
<heat> currently my IO is fully synchronous, i.e you can't tell it "fullfill this request and tell me when it's done"
<heat> this is an attempt to improve that situation
<heat> because that obviously sucks
<heat> ofc io_uring would depend on this queue work but i'm really far from adding that lol
<moon-child> oh so it's just general async io
<moon-child> that's still good
<heat> yep
<heat> i also need something similar for e.g NIC packet queues
xenos1984 has quit [Read error: Connection reset by peer]
<heat> basically i went around and made everything synchronous in a lot of hardware when the optimal and more accurate way to model things would be with fully asynchronous, queued operation
heat has quit [Remote host closed the connection]
goliath has quit [Quit: SIGSEGV]
cross has joined #osdev
kerravon has joined #osdev
xenos1984 has joined #osdev
<zid> fun, the heavens just opened out of nowhere
<zid> silence, and then so much rain I can't hear anything else
<gog> neat
<zid> aand it's gone
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<zid> rain is weird
<gog> microburst
heat has joined #osdev
<heat> holy shit thank god for system restore
<zid> I've never used it
<heat> who needs to die in linux desktop so we get this feature
<zid> it's called "rsync /etc somewhere every day"
<heat> so this w11 update was busted and fucked the system up
<zid> some people keep everything in /etc in a git repo
<heat> well, surely rsync /etc /usr/bin /usr/lib*
<zid> meh you can just reinstall those
<zid> from tgzes
<heat> well, the point is that the binaries got fucked
<heat> and i had a way out
<zid> like if my gentoo system exploded, I could just start with a stage3, copy /etc over (including my @world set) and hit emerge
<zid> and everything would be back to normal
eddof13 has joined #osdev
eddof13 has quit [Client Quit]
<gog> i always keep a backup linux usb
<gog> no muss no fuss no relying on a software solution built into the system that could itself fail
<zid> except uefi
<zid> thankfully uefi never fails
<zid> right heat?
<heat> nuh huh never
<gog> uefi is the most stable and sensible solution ever
<heat> fault-proof
<gog> it's absolutely perfect in every way
<zid> I still laugh, heat asked me one day why I used bios boot not uefi
<zid> and I replied "I dont give a shit about secure boot and uefi only adds places for it to go wron"
<zid> heat was vaguely dismissive
<zid> then he blows up his uefi and I get to laugh myself to sleep
<gog> you have to secure your boots
<heat> well, no
<heat> your bios boot was still uefi
<zid> and it broke too
<kazinsal> boot status: secure
eddof13 has joined #osdev
heat has quit [Quit: Client closed]
<gog> boot status: doc martens
<gog> ¯\_(ツ)_/¯
<gog> extensible footwear interface (my feet)
<zid> uefi -> gog sends you feet pics if you ask
<kazinsal> an unexpected kind of boot image
<gog> if you ask really nicely i'll step on you
<kazinsal> *goodness*
heat has joined #osdev
<heat> gosh i fucking hate computers
<heat> god's curse upon humanity
<gog> yes
<gog> it is pandora's box
<zid> Computers are okay, it's when people try to pretend they're useful that it's a proble
<gog> it contains the entirety of human joy and despair
<gog> turns out there's a fuckload more despair
<zid> they write fancy UIs and things instead of giving 14 reams of diagnostic info
<kazinsal> 99 terabytes of despair, 1 terabyte of feet pics
<gog> yes
<zid> It should cost at least $7 in ink to boot windows
<heat> ok lets bisect fellas
<heat> ...is fellas a gender neutral term?
<kazinsal> I'll grab the chainsaw
<zid> yes, so is chaps
<zid> and mates and blokes and bitches
<kazinsal> and me ol' cocks
<heat> ok so its not
<heat> thanks
gorgonical has joined #osdev
<gog> getting bisected in a sundress
<gog> ¯\_(ツ)_/¯
<zid> pervert
<kazinsal> chat kinda sus tonight, as the kids say
<gog> sorry
<zid> <gog>GETTING FUCKED REALLY HARD IN A SKIRT
<gog> i didn't say that
<zid> 100% you said that
<gog> you said it
<gog> i said bisected
<heat> linux kernel operaing system
<gog> yes
<bslsk05> ​preview.redd.it: Reddit - Dive into anything
<heat> ok found the bad commit
<heat> these git fucks
<heat> they know what they're doing
Turn_Left has quit [Read error: Connection reset by peer]
Burgundy has quit [Ping timeout: 272 seconds]
<heat> woah, found it
<gog> yay
<zid> what were you even looking for
<heat> a big fork regression
<heat> i added "marks" to my radix tree which is kind of a tag system that gets propagated up and down the table. one of them is super important, the present tag, that allows me to traverse the whole table with bitops vs reading 64 * sizeof(long) bytes
<heat> but i forgot to copy them when copying a whole radix tree...
<heat> so forked vm objects (that use radix trees to store pages) would show up as empty and would re-fault a new page
<heat> i should add a pre-push hook for CI
dude12312414 has joined #osdev
pounce has quit [Changing host]
pounce has joined #osdev
dude12312414 has quit [Remote host closed the connection]
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
vai has quit [Ping timeout: 258 seconds]
SGautam has quit [Quit: Connection closed for inactivity]
<heat> mjg: atm my page_fault1 is the definition of "doesn't scale"
<heat> as in it -t4 in threads = -t1 in threads
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
mhall has quit [Quit: Connection closed for inactivity]
nyah has quit [Ping timeout: 264 seconds]