Arthuria has quit [Remote host closed the connection]
netbsduser`` has joined #osdev
netbsduser`` has quit [Ping timeout: 258 seconds]
<kof213>
https://dmitry.gr/?r=05.Projects&proj=25.%20VMU%20Hacking > It is a pity that the CPU architecture is so user-hostile that I cannot imagine targeting a C compiler to it. Nobody's going to write serious code in assembly for an obscure old CPU, right? > Well...I would. > Currently no Cortex-M23 silicon exists, so the VMU now might be the first real device you can get your hands on that will run Cortex-M23 code! Ha!
<kof213>
> Not only does it expose all of the VMU's functionality to the emulated code, it even allows interrupt handling, nesting, and exceptions to work properly like a real Cortex-M23
netbsduser`` has joined #osdev
netbsduser`` has quit [Ping timeout: 240 seconds]
rustyy has quit [Read error: Connection reset by peer]
netbsduser`` has joined #osdev
rustyy has joined #osdev
Vercas has quit [Quit: Ping timeout (120 seconds)]
netbsduser`` has quit [Ping timeout: 255 seconds]
goliath has joined #osdev
Vercas has joined #osdev
netbsduser`` has joined #osdev
gxt has quit [Remote host closed the connection]
gxt has joined #osdev
netbsduser`` has quit [Ping timeout: 255 seconds]
danilogondolfo has joined #osdev
Left_Turn has joined #osdev
Turn_Left has joined #osdev
Left_Turn has quit [Ping timeout: 264 seconds]
blop_ has quit [Ping timeout: 252 seconds]
blop_ has joined #osdev
sauce has joined #osdev
<zid>
can we free(cmake) after
GeDaMo has joined #osdev
pg12 has quit [Remote host closed the connection]
pg12 has joined #osdev
pg12 has quit [Remote host closed the connection]
pg12 has joined #osdev
pg12 has quit [Remote host closed the connection]
pg12 has joined #osdev
gog has joined #osdev
gbowne1 has quit [Quit: Leaving]
pg12 has quit [Remote host closed the connection]
pg12 has joined #osdev
pg12 has quit [Remote host closed the connection]
pg12 has joined #osdev
pg12 has quit [Remote host closed the connection]
pg12 has joined #osdev
Burgundy has joined #osdev
Osmten has joined #osdev
viatatribal has quit [Quit: Leaving]
netbsduser`` has joined #osdev
netbsduser`` has quit [Read error: Connection reset by peer]
nyah has joined #osdev
jjuran has quit [Ping timeout: 255 seconds]
Vercas has quit [Ping timeout: 252 seconds]
Vercas has joined #osdev
Vercas has quit [Remote host closed the connection]
Vercas has joined #osdev
bauen1 has quit [Ping timeout: 240 seconds]
[itchyjunk] has joined #osdev
Yoofie has quit [Ping timeout: 255 seconds]
Osmten has quit [Quit: Client closed]
Burgundy has quit [Ping timeout: 248 seconds]
bauen1 has joined #osdev
Burgundy has joined #osdev
<sham1>
::operation delete cmake;
<zid>
I am gas, gas, gas
<mcrod>
hi
<gog>
mewo
<mcrod>
gog may I hug you
<Ermine>
hi gog, may I pet you
<gog>
mcrod: yes
<gog>
Ermine: yes
* mcrod
hugs gog
* gog
hug
* Ermine
pets gog
* gog
prr
admiral_frost has joined #osdev
Matt|home has quit [Read error: Connection reset by peer]
Matt|home has joined #osdev
goliath has quit [Quit: SIGSEGV]
heat has joined #osdev
<mcrod>
unrelated
<mcrod>
i actually feel a little horrified at how much employers seem to be spying on github accounts
<mcrod>
i don't think this is happening widespread
<zid>
spy on their wives
<mcrod>
but a friend of mine got shit on earlier this morning because he made a commit to one of his repos, and his boss messaged him about 2 hours later asking "weren't you supposed to be working"
<Cindy>
mcrod: "you worked on something on your sick day?"
<mcrod>
I wish it was practical to be completely, fully anonymous sometimes
<mcrod>
but... that's harder than ever
<Cindy>
"you like working on your open source crap more than here? then get out"
<mcrod>
i tried to be fully anonymous a year ago
<mcrod>
had a list of things on my desk not to say or do
<mcrod>
it was impossible
<mcrod>
everything blocks tor, everything shits on proxies, can't use CAPTCHAs, can't use discord
<mcrod>
using github was virtually impossible, although I could commit and push through tor
<mcrod>
and proving to employers that you are who you say you are became much harder
<bslsk05>
EtchedPixels/FUZIX - FuzixOS: Because Small Is Beautiful (250 forks/2016 stargazers/NOASSERTION)
<heat>
alan cox has his own sysv clone
<kof213>
yes, i am not a "privacy is gone, get over it" person, but at the same time...what can you do without littering details all over the place
<kof213>
sit in the woods at a typewriter
<kof213>
i'm sure the modern typewriters would have chipped ink lol
<GeDaMo>
Just move to the other Internet
<kof213>
or, you have a family member/friend someone pay all your bills. ok, then they have no privacy lol
<kof213>
assuming you could convince them in the first place.
<kof213>
and you trust everything in their name...that could never backfire
<kof213>
> Just move to the other Internet that just makes me think john wick bowery king pigeonnet
<zid>
can we use the normal internet
<zid>
but move to a different web
<zid>
I don't wanna have to have two modems
Matt|home has quit [Ping timeout: 248 seconds]
MaxLeiter has quit [Remote host closed the connection]
MaxLeiter has joined #osdev
<zid>
I have carbs on carbs for dinner, I win
<Bitweasil>
I know more and more tech sorts who are just... over consumer tech in their personal lives.
<Bitweasil>
I'm pretty much there. I'd rather spend my time around an old truck or something than screwing with computers.
<Bitweasil>
Gave up on smartphones, went back to a flip device, I don't carry it often, etc.
<Bitweasil>
Qubes + Tor solves a lot, but as mcrod noted, you can't access a lot of the internet through it. Even various read only sites are blocking Tor exit nodes for "We won't tell you, but you know who you are..." sort of nonsense reasons.
xenos1984 has quit [Ping timeout: 240 seconds]
kfv has joined #osdev
xenos1984 has joined #osdev
<mcrod>
it’s sad
<zid>
never encountered a tor block that wasn't "your ip is banned cus abuse" and a quick alt-shift-l didn't fix
<mcrod>
Bitweasil: i think for me, it became nothing but a profit machine in 10 years
<mcrod>
there's no joy in anything anymore
<mcrod>
i'm.. really not sure how to put my discontent into words
<zid>
I have an imgur gallery you can read that may resonate
<bslsk05>
randomascii.wordpress.com: 32 MiB Working Sets on a 64 GiB machine | Random ASCII – tech blog of Bruce Dawson
<Bitweasil>
zid, alt-shift-l?
<gog>
return to monke
<heat>
monke unix
<gog>
my boss broke the fucking build again
<gog>
god damn it
<heat>
perdon gog
<heat>
por veces acontece si?
<gog>
no, el jefe le rompé construcción de la programa
<Bitweasil>
zid, ouch. That gallery hits pretty hard. Not wrong, though.
<heat>
a el jefe le encanta la construcción del núcleo linux
<netbsduser>
i bet the jeffe really is enchanted with the construction of the nucleus of linux
<netbsduser>
i would be too
<gog>
nadie se encantan el núcleo linux
<heat>
vamonos usar un núcleo UNIX más humilde
<gog>
claro
<heat>
con grandes funcións como lookuppn
wgrant has joined #osdev
<heat>
y pagedaemon
<heat>
dem
<heat>
demonio de páginas
<gog>
que es 'página'? regresemos al segmentatción
<netbsduser>
that phrase just doesn't seem quite write when translated on another language
<gog>
regresmos al mona
<netbsduser>
"duilleag diabhal"
<gog>
i should go home
<heat>
no gog, yo quiero usar el VAX con páginas de 512 mordidas
<heat>
porque eres el tamaño de un sector
<gog>
sí, no mas con sectors de 4096 mordidas
<gog>
"formato avanza" mi culo
xenos1984 has quit [Ping timeout: 252 seconds]
<heat>
yo quiero un formato de paginación con listas enlazadas
<heat>
como ce dice en ingles, eres mucho OPTIMAL
<gog>
en ingels es "PERFORMANT"
<Ermine>
Что?
<heat>
friedrich ingels
<heat>
jajajajajajajaja
<zid>
can you write the portuguese bits in a language someone has actually heard of, like church latin
<zid>
or moldovan
<GeDaMo>
Klingon
<heat>
nu am vorbit limba portugheza zid
<heat>
also wow the google translate latin support kinda sucks ass
<zid>
google translate sucks for language
<zid>
which is unfortunate
xenos1984 has joined #osdev
gog has quit [Quit: Konversation terminated!]
<heat>
so what's it good at? moving objects along an axis?
sham1 is now known as sham1`
sham1 has joined #osdev
sham1` has quit [Quit: WeeChat 3.7.1]
lockna has joined #osdev
kfv has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
lockna has quit [Quit: Leaving]
<heat>
"Like the Zen/Zen+ microarchitecture, Zen 2 supports page table entry (PTE) coalescing. When the table walker loads a PTE, which occupies 8 bytes in the x86-64 architecture, from memory it also examines the other PTEs in the same 64-byte cache line. If a 16-Kbyte aligned block of four consecutive 4-Kbyte pages are also consecutive and 16-Kbyte aligned in physical address space and have identical page attributes, they are stored into a single TLB entry
<heat>
greatly improving the efficiency of this cache"
<heat>
zen is just an arm64 core with a different frontend and it's very funny
kfv has joined #osdev
<Bitweasil>
Yeah, there's no "contiguous" bit, and I think ARM prefers 8-entry contiguous blocks, but... very similar. :)
<heat>
hmmmmm i am wondering
<heat>
does it set the A bit?
<heat>
A should only be set if the instruction retires though...
<Bitweasil>
The accessed flag?
<Bitweasil>
The ARMv8 manual explicitly says that speculative accesses can set it.
<Bitweasil>
("may")
bauen1 has joined #osdev
<heat>
i don't think x86 does though? although may be wrong
<Bitweasil>
Yeah, but x86 is really strictly formal with memory accesses/ordering/etc compared to ARM.
<bslsk05>
lobste.rs: 32 MiB Working Sets on a 64 GiB machine | Lobsters
zid has quit [Remote host closed the connection]
gog has joined #osdev
admiral_frost has quit [Quit: It's time]
zid has joined #osdev
gog has quit [Ping timeout: 252 seconds]
<geist>
heat: i'm guessing the AMD coalesces it but maybe remembers which entry caused the TLB load so it writes back to just that one?
<geist>
or maybe A bit is sloppy. but D bit has to be very precise
SGautam has joined #osdev
kfv has quit [Ping timeout: 272 seconds]
kfv has joined #osdev
gbowne1 has joined #osdev
goliath has joined #osdev
antranigv has joined #osdev
<Bitweasil>
I don't think it does...
<Bitweasil>
Oh, I think you're right. The architecture does not permit updates to AP[2] by the hardware management of dirty state to occur as a result of speculative accesses.
<Bitweasil>
But speculative accesses can update the access flag.
<geist>
right. think of the A bit writeback as happening asynchronously when the cpu first fetches the TLB entry from the page table
<geist>
as it loads it, it notices the A bit isn't set so it writes it back then. hence why speculative accesses may update it
<Bitweasil>
Makes sense.
<geist>
but D bit is more precise, you have to write it back only if you actually commit a write, so speculative writes have to stay in a queue (which they do anyway) and only update if written back
<Bitweasil>
We don't speculate, so I can ignore some of that, but I did have to look up the access flag behavior recently.
<geist>
FWIW as verbose as the ARM manual is, once you absorb it you really learn a lot about how modern architetectures work, since most of these techniques they describe are universal
<Bitweasil>
Yeah, though I think the ARMv8 manual is both needlessly verbose in some areas and "scatters the relevant information over 17 separate sections."
<geist>
yah
<geist>
thankfully for things like a random register, their naming convention is solid enough that you can just search for it
<geist>
TCR_EL1, etc
<geist>
and in general you can find the first mention, and then that's hyperlinked to the definition of it, etc
<geist>
so it's pretty good for random lookup
<Bitweasil>
Mostly. But, yes. They do hyperlink within the manual very well.
<geist>
provided you have a supercomputer that can hold the PDF in memory and process it in a reasonable amount of time
<geist>
i wonder if i took like a llama2 model and trained it on these manuals if it'd actually be useful
<geist>
i fired up llama.cpp for the first time last night, was kinda interesting, actually works
<geist>
wasn't *too* bad. a 13B model took about 10GB ram and lit up 16 cores of my 3950x and would produce a word about every 250ms
<geist>
so was reasonably fast
eddof13 has joined #osdev
kfv has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
GeDaMo has quit [Quit: That's it, you people have stood in my way long enough! I'm going to clown college!]
<relipse>
does macos contain any linux?
<Bitweasil>
I don't think it ever had any.
<Bitweasil>
It was always BSD based.
<heat>
sure it does
<heat>
GNU bash
<heat>
if you count that as linux
<puck>
BSD stands for Bash S....okay i can't think of anything funny
<heat>
GNU SLASH LINUX, OR AS I'VE RECENTLY TAKEN TO CALL IT, GNU PLUS LINUX
<puck>
GNU + MACOS
<heat>
GNU PLUS NETBSD PLUS FREEBSD
<heat>
geist, yeah im just curious cuz I use the A bit to shootdown TLB entries
<heat>
which is honestly a questionable optimization but whatever
<heat>
>Because SQL Server on Linux is SQL Server on Windows, with the NT kernel running as a user-mode Linux process. See Drawbridge.
<heat>
fuckin
<heat>
eyebleach
<heat>
i need some
<geist>
huh now that's fascinating
<geist>
or... probably m ore correctly, it's some sort of NT emulation shim, and the quote is not really correct
<geist>
since most folks can't tell the precise difference between a kernel and something that looks like a kernel
<bslsk05>
arstechnica.com: How an old Drawbridge helped Microsoft bring SQL Server to Linux | Ars Technica
<heat>
it seems kinda like a libos
* mcrod
vomit
<mcrod>
MSSQL
<Ermine>
So they brought windows along with ms sql to linux
<heat>
finally
<heat>
that's whta i was looking for
<heat>
windows on linux
netbsduser has quit [Ping timeout: 240 seconds]
<Mondenkind>
hrm. A subsystem for linux, to let it run windows applications--a windows subsystem, for linux?
<targetdisk>
so question. does the Posix-UEFI library require that I compile for strict ANSI C or can I use newer C2X features/extensions?
gog has joined #osdev
<geist>
heat: yeah so a libos doesn't really seem that terrible. that's done all the time, build a shim to translate one to the other
<geist>
targetdisk: ugh, that's lame. honestly i really dont like that library. if i were to uefi stuff i'd probably just write my own, or steal from fuchsia
<geist>
but, i'd probably only be using a limited amount of uefi anyway
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
netbsduser has joined #osdev
<targetdisk>
wait so can I compile in C99 mode or not tho?
<targetdisk>
GNU-EFI is the lame one. I found POSIX-UEFI okay
<heat>
what
<heat>
POSIX-UEFI is garbie
<gog>
clang
<gog>
heat: clang
<heat>
"what if UEFI, but POSIX" is like the most overkill shit i've ever heard
<heat>
also the author is a PoS IIRC
<heat>
gog, msvc
<gog>
i'm implementing POSIX in UEFI heat
<heat>
no
<heat>
dont
<gog>
i am
<heat>
you are not
<gog>
did you see my janky syscall abuse
<heat>
i like you
<heat>
don't make that change
<heat>
i can't have deep conversations with people that implement POSIX in UEFI
<heat>
those people aren't to be trusted
<mcrod>
hi
<gog>
mcrod is helping me
<gog>
tell him
<mcrod>
heat is friend
<mcrod>
but
* mcrod
point and laugh at heat
<heat>
oh yeah it's bzt
<heat>
fuck bzt
<gog>
who is bzt
<Ermine>
author of gnu-efi afaik
<Ermine>
and bootboot
<heat>
a guy who flamed around in the osdev forums for a good while
<heat>
no, author of posix-uefi
<Ermine>
and scalable fonts thing
<kazinsal>
for a while he was banned from the forums
<kazinsal>
someone let him back in
<kazinsal>
idk why
<kazinsal>
I think he was banned from the freenode channel as well
<mcrod>
when I hear "GNU"
<gog>
i'm going to have a whole POSIX-compliant environment in UEFI
<mcrod>
I think "I'm 40,000 years old and I'm stuck in my old ways"
<kazinsal>
unix as a uefi application
<kazinsal>
whole v7 filesystem implemented on top of FAT32
<Ermine>
and now he claims that toaruOS is the only OS allowed for discussions and worshipping here
<heat>
mcrod, you're literally using GNU software atm
<mcrod>
yes
<mcrod>
but I also built gcc recently
<mcrod>
and I read the mailing lists
<heat>
building gcc is easy
<mcrod>
it is impossible to imagine that they function on a regular basis
<heat>
LLVM is so much worse with its bespoke build system in tons of arcane cmake variables
<mcrod>
I'm starting to think I went overkill with this project though
<mcrod>
however, I will see it through to the end
<heat>
like yes gcc is annoying to tinker with if you have to mess with autoconf
<heat>
however, LLVM's build system is a huge LUL
<Ermine>
Why did they even chose cmake
<heat>
it's the 'default' C++ build system
<mcrod>
there is a bug in the CMake scripts though
<mcrod>
and I spent too long on #llvm
<kazinsal>
heh. that posix-uefi repo's readme opens with a screed about how reddit is witch hunting him because he needs to use a warning disable flag to stop the compiler from complaining that he's doing some kind of on the fly transparent UTF-8 to UTF-16 conversion in redefined builtins, and the compiler is going "wtf are you doing"
<mcrod>
only for someone to tell me "yes, it sucks, I wrote it and I'm telling you it sucks"