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
<mrvn> do you have ARM servers?
Burgundy has left #osdev [#osdev]
gildasio has quit [Ping timeout: 268 seconds]
FreeFull has quit []
<gog> only LEG
pretty_dumm_guy has quit [Quit: WeeChat 3.5]
arminweigl_ has joined #osdev
gildasio has joined #osdev
arminweigl has quit [Ping timeout: 240 seconds]
arminweigl_ is now known as arminweigl
lg has quit [Remote host closed the connection]
sprock has quit [Ping timeout: 245 seconds]
sprock has joined #osdev
vdamewood has quit [Ping timeout: 240 seconds]
vdamewood has joined #osdev
gog has quit [Ping timeout: 252 seconds]
dude12312414 has joined #osdev
[itchyjunk] has quit [Ping timeout: 240 seconds]
[itchyjunk] has joined #osdev
matrice64 has quit [Quit: Textual IRC Client: www.textualapp.com]
[itchyjunk] has quit [Remote host closed the connection]
hypoalgesia has quit [Quit: WeeChat 3.6]
poyking16 has joined #osdev
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
Matt|home has joined #osdev
wootehfoot has joined #osdev
wootehfoot has quit [Ping timeout: 268 seconds]
MiningMarsh has quit [Quit: ZNC 1.8.2 - https://znc.in]
gildasio has quit [Ping timeout: 268 seconds]
gildasio has joined #osdev
MiningMarsh has joined #osdev
dude12312414 has quit [Quit: THE RAM IS TOO DAMN HIGH]
opal has quit [Remote host closed the connection]
gildasio has quit [Remote host closed the connection]
foudfou has quit [Remote host closed the connection]
gildasio has joined #osdev
opal has joined #osdev
foudfou has joined #osdev
poyking16 has quit [Read error: Connection reset by peer]
poyking16 has joined #osdev
Oshawott has quit [*.net *.split]
ids1024 has quit [*.net *.split]
dormito has quit [*.net *.split]
Geertiebear has quit [*.net *.split]
friedy has quit [*.net *.split]
mrkajetanp has quit [*.net *.split]
Oshawott has joined #osdev
mrkajetanp has joined #osdev
dormito has joined #osdev
foudfou has quit [Remote host closed the connection]
foudfou has joined #osdev
the_lanetly_052 has joined #osdev
dragestil has quit [*.net *.split]
afarrag has quit [*.net *.split]
geist has quit [*.net *.split]
phr3ak has quit [*.net *.split]
bslsk05 has quit [*.net *.split]
eschaton has quit [*.net *.split]
corecode has quit [*.net *.split]
varad has quit [*.net *.split]
Emil has quit [*.net *.split]
merry has quit [*.net *.split]
afarrag has joined #osdev
geist has joined #osdev
Emil has joined #osdev
varad has joined #osdev
corecode has joined #osdev
dragestil has joined #osdev
phr3ak has joined #osdev
poyking16 has quit [Ping timeout: 244 seconds]
eschaton has joined #osdev
merry has joined #osdev
poyking16 has joined #osdev
psykose has quit [Remote host closed the connection]
psykose has joined #osdev
arminweigl_ has joined #osdev
arminweigl has quit [Ping timeout: 252 seconds]
arminweigl_ is now known as arminweigl
bauen1 has quit [Ping timeout: 252 seconds]
foudfou_ has joined #osdev
foudfou has quit [Ping timeout: 268 seconds]
opal has quit [Remote host closed the connection]
opal has joined #osdev
gildasio has quit [Remote host closed the connection]
gildasio has joined #osdev
epicusername has joined #osdev
epicusername has quit [Quit: Igloo IRC: https://iglooirc.com]
SpikeHeron has quit [Ping timeout: 252 seconds]
SpikeHeron has joined #osdev
curi0 has joined #osdev
curi0 has quit [Quit: Leaving]
scaleww has joined #osdev
wootehfoot has joined #osdev
wootehfoot has quit [Ping timeout: 240 seconds]
zaquest has quit [Remote host closed the connection]
bauen1 has joined #osdev
zaquest has joined #osdev
GeDaMo has joined #osdev
the_lanetly_052_ has joined #osdev
socksonme_ has joined #osdev
the_lanetly_052 has quit [Ping timeout: 252 seconds]
bslsk05 has joined #osdev
orccoin has joined #osdev
heat has joined #osdev
sprock has quit [Ping timeout: 268 seconds]
gxt__ has quit [Write error: Connection reset by peer]
foudfou_ has quit [Remote host closed the connection]
gxt__ has joined #osdev
foudfou has joined #osdev
sprock has joined #osdev
opal has quit [Remote host closed the connection]
opal has joined #osdev
lkurusa has joined #osdev
gog has joined #osdev
the_lanetly_052_ has quit [Ping timeout: 245 seconds]
foudfou has quit [Remote host closed the connection]
foudfou has joined #osdev
gxt__ has quit [Remote host closed the connection]
gxt__ has joined #osdev
SpikeHeron has quit [Quit: WeeChat 3.5]
opal has quit [Ping timeout: 268 seconds]
SpikeHeron has joined #osdev
opal has joined #osdev
nur has quit [Remote host closed the connection]
scaleww has quit [Read error: Connection reset by peer]
poyking16 has quit [Ping timeout: 255 seconds]
[itchyjunk] has joined #osdev
heat_ has joined #osdev
poyking16 has joined #osdev
heat has quit [Read error: Connection reset by peer]
poyking16 has quit [Ping timeout: 240 seconds]
poyking16 has joined #osdev
PapaFrog has quit [Remote host closed the connection]
PapaFrog has joined #osdev
poyking16 has quit [Quit: WeeChat 3.6]
poyking16 has joined #osdev
<Bitweasil> mjg, they tend to have various forms of them, but far less severe than Intel, typically.
<Bitweasil> And there are still some fairly modern ARM cores that don't speculate or don't speculate far, so you can use them without much risk.
<Bitweasil> I believe the A53s are still in order.
lkurusa has quit [Quit: I probably fell asleep (or went out). Who will ever know.]
<geist> yah for the most part every time a new spectre style thing comes along the cortex-a3x and cortex-a5x series are immune
<geist> mostly because they are largely in order machines. think of classic N stage in order pipeline designs from the late 80s
<zid> actual riscs? :o
<geist> the 5x and whatnot series are designed to sit in a particular spot on the power/performance chart
<jimbzy> I really need to finish my antenna today.
<heat_> can confirm
heat_ is now known as heat
<heat> my a53 rpi has no mitigations
<jimbzy> Wrong channel, but it still needs to be done! :p
<zid> cantenna is good enough for anyone
<jimbzy> Nah, I'm building a 70cm yagi
<zid> You're going to struggle to find pringles that big
pretty_dumm_guy has joined #osdev
<zid> maybe you could use lots of them
<jimbzy> I'm actually making it out of PVC and an old tape measure for the elements XD
<gog> 70cm is like two pringles cans
<jimbzy> My little 2ft vertical can pick up the ISS when it's overhead, but I can only hear it for about a minute.
<jimbzy> Once I can hear it better then I'll build a 2m so I can transmit.
<zid> gog: i was thinking 70cm dia/radius
<jimbzy> Probably use copper tubing for that one.
<zid> so you'd need to make a big circle of cans
<zid> lashed together
<gog> ohhhh
<gog> yeh
<jimbzy> Nah. 70cm radio band, 440mhz-ish
<jimbzy> ISS is 437.800mhz downlink and 14x.xxxmhz for the uplink.
<zid> so yes you'd need a 70cm pringles tube
<jimbzy> Man, now I'm craving pringles...
<zid> pringles ended up being just about the right size for 2.4GHz wifi
<bslsk05> ​www.instructables.com: The Tape Measure Antenna : 5 Steps (with Pictures) - Instructables
<jimbzy> Basically building something like that.
wootehfoot has joined #osdev
<zid> Like I said, you're going to need a super big pringles can to put that in :p
<jimbzy> Then, I'm going to sit on the roof of my house with my radio screaming at the satellites.
<zid> I just open the door of the microwave
<zid> The only message I can transmit is "Cooking..." though
<jimbzy> Hehehe
bauen1 has quit [Ping timeout: 240 seconds]
<gog> i wanna transmit AAAAAAAAAAAAAAAAAAAAa to the stars
<zid> I wonder if anyone has ever ddos'd a sat
<gog> probably
<jimbzy> You can use too much power and block other users.
<gog> yeh that's basically the easiest way, just drown out the signal with noise on the same band
<zid> I assume to the sat it'd look a bit like a laser pen vs plane cockpit
<zid> and it'd struggle to see the blinkenlights from other people
<jimbzy> Yeah. A lot of people start transmitting before they've captured it and it creates a lot of funky noise.
<zid> So I wonder if anyone has ever done it on purpose, like some company shafted them so they just pointed a transmitter at it all day
<jimbzy> That's why I don't even try at this point. I'm just listening right now.
<zid> I assume in the US the FCC would end up knocking on your door
<jimbzy> Yesterday there was a young lady working it like a boss. A 13 year old from MN.
<jimbzy> Yeah
brynet has quit [Ping timeout: 255 seconds]
<jimbzy> Fun hobby, though.
vai has joined #osdev
<vai> hi
<gog> hi
<jimbzy> Howdy
<zid> no
<gog> witam
<jimbzy> No u
<gog> eee
<zid> gog give me one of your finest eths please
<gog> lol i don't own crypto
<zid> wat
<zid> fine I'll wikipedia my own
<zid> ðausages.
<jimbzy> My older brother asked me about crypto the other day...
<zid> I have a weird lisp
<gog> þausages
<zid> gog is speaking crypto
<jimbzy> I told him to give me half of what he was thinking about investing and in return I'd punch him in the stomach as hard as I could, and he'd still come out ahead.
<gog> oh you wanted the letter eth
<gog> ðððððð
<zid> I already got my own!
<gog> why aren't you using the compose key
<zid> because the default windows layout has fucking nothing good
<gog> altgr-d-h
<zid> only é
<gog> oh
<gog> hahaha windows
<zid> every other key is dead
<zid> and I'm too lazy to add a bunch of cool shit like yen and pi
<zid> Ðat would be a lot of effort
<gog> það
<GeDaMo> That reminds me, I was trying to add a compose for lambda
<jimbzy> I booted up windows 10 the other night for the first time and all it did was mess up my UEFI settings...
<kingoffrance> </t-shirt>
<zid> wikipedia says icelandic can't start words with Ð
<jimbzy> kingoffrance, Yep
<zid> english doesn't care, english superior
<zid> sorry, forgot about my lisp, engliþ
<gog> not that it can't, just that it doesn't in the modern orthography
orccoin has quit [Ping timeout: 240 seconds]
<zid> it's only orthography? not phonotactics? how do they spell it then?
<jimbzy> With letters?
<zid> yes, but you can't use the right one, apparently
<zid> I *assume* we actually mean phonotactics
<zid> but not many people know that word
<gog> id
<gog> idk
<zid> Basically it's like.. english has a 'ng' sound but only uses it at the end of words, even though it's perfectly possible to, see: vietnamese
<jimbzy> I'm only a 3rd Technician in the Space Corps. :(
<zid> aren't you supposed to be unclogging the nozzles of the chicken soup dispensers
<zid> That rule is phonotactics, syntax for using sounds
<jimbzy> :]
<jimbzy> Gazpacho soup...
<zid> so basically, icelandic is silly, as some of ðe best english words start with a ð
<gog> icelandic is very silly
<zid> My favourite is that germans can't say þs
<zid> Listening to them say the word "tenths" in racing game is always popcorn worthy
<gog> tents
<zid> tentusuthustth
Raito_Bezarius has quit [Ping timeout: 240 seconds]
Raito_Bezarius has joined #osdev
brynet has joined #osdev
rwb has quit [Ping timeout: 252 seconds]
pretty_dumm_guy has quit [Ping timeout: 240 seconds]
brynet has quit [Ping timeout: 255 seconds]
pretty_dumm_guy has joined #osdev
rwb has joined #osdev
vai has quit [Quit: Lost terminal]
SpikeHeron has quit [Quit: WeeChat 3.5]
SpikeHeron has joined #osdev
<mats1> cute
rwb has quit [Ping timeout: 245 seconds]
rwb has joined #osdev
bauen1 has joined #osdev
weinholt has quit [Read error: Connection reset by peer]
rwb has quit [Ping timeout: 245 seconds]
rwb has joined #osdev
<gorgonical> Anyone know how syslinux decides what a valid image magic is?
<zid> will you be upset if I say 'magic'
<gorgonical> Only a little
<zid> Magic.
wootehfoot has quit [Ping timeout: 245 seconds]
foudfou_ has joined #osdev
foudfou has quit [Ping timeout: 268 seconds]
eck has quit [Quit: PIRCH98:WIN 95/98/WIN NT:1.0 (build 1.0.1.1190)]
eck has joined #osdev
<dzwdz> can grub ever fail to provide a linear framebuffer on x64?
<zid> I mean, sure, headless machines
<zid> with only basic EGA or whatever
<heat> any machine without a gpu
<zid> or none at all
<heat> nothing guarantees your BIOS or UEFI calls to the video driver actually have a backend
<heat> in UEFI it's managed by you looking for the protocol - if no one installed that, EFI_NOT_FOUND
poyking16 has quit [Ping timeout: 268 seconds]
gildasio has quit [Remote host closed the connection]
gxt__ has quit [Write error: Connection reset by peer]
gildasio has joined #osdev
gxt__ has joined #osdev
brynet has joined #osdev
heat has quit [Remote host closed the connection]
heat has joined #osdev
<bslsk05> ​arstechnica.com: Linus Torvalds uses an Arm-powered M2 MacBook Air to release latest Linux kernel | Ars Technica
<zid> saw marcan earlier mentioning he's found a lot of bugs in various bits of software trying to get asahi linux running on m1/m2
<zid> because nobody else builds desktop software on non-x86
<zid> until m1 came around
<zid> so lots of things like chrome etc use jemalloc which doesn't understand non-4k pages etc
<mjg> heh
<heat> arm64 usually uses 4KB pages
<mjg> they roll with 16k?
<zid> there is no arm64 'usually' for desktop software though
<zid> that's the point
<heat> mjg, yes
<mjg> is there any write up how that's working out?
<mjg> memory waste-wise
<heat> dunno
<zid> 16k's probably what you want for desktop apps
<zid> given how everything is just managed by malloc afterwards
<zid> and you don't have to care about your 4MB program not aligning to your page size well at load time
<mjg> well i'm worried that malloc wont be able to release much memory
<j`ey> mjg: dont think anyone did some general analysis
<j`ey> mjg: someone did some kernel compilation benchmarking with different page sizes
<mjg> unless there is some coop where they can demote these pages?
<heat> j`ey, how much faster are the 16KB pages?
<zid> seven fasters
<j`ey> lemme try find the details
<heat> go go arm man
<heat> sorry, aarch64
<bslsk05> ​redirect -> docs.google.com: compile speed by page size and available ram - Google Sheets
<j`ey> zid: jemalloc worked fine on 16K pages.. it's just that it hardcoded page size at build time
<mjg> heh kernels per hour
<j`ey> so if you rebuild it, it's fine
<mjg> nice metric
<zid> j`ey: I think probably chrome etc just never bothered to hook the defines up to anything sensible
<j`ey> im sticking with 16K.. because why not
<j`ey> zid: someone fixed chromium now at least
<j`ey> I mean it already supported 16K because of iOS
<zid> What's the prefered method for doing that anyway
<zid> I can't think it's to include kernel headers
<j`ey> well it's dynamic
<zid> well that solves the issue then more or less
<heat> zid, for doing what?
<zid> now you have to do it dynamically
<j`ey> yup
<bslsk05> ​github.com: chromium/page_allocator_constants.h at main · chromium/chromium · GitHub
<zid> are they aware that they have code in their .h file, smh
<j`ey> what else goes in .h files? just comments? :P
<zid> declarations
<zid> and ascii art
<zid> can't forget the ascii art
<j`ey> tbh not a lot of stuff broke.. but chromium was a big one, especially since it affected electron too
wootehfoot has joined #osdev
<zid> I bet a lot of stuff broke in the grand scheme of "anything you might wanna install via distro", but not proportionally
<heat> is linux actually usable on the m1?
<j`ey> define usable? :P
<zid> can you watch porn
<heat> there's no GPU accel
<heat> exactl
<heat> y
<j`ey> no GPU accel yet, but that's being worked on
<heat> that will take like 5 years
<j`ey> but wifi works, bluetooth
<heat> and 5 more for actually stable acceleration
<heat> or 10 really
<zid> porn is hard, needs wifi, accelerated crypto, chrome, x264, hardware gpu accel, kms, etc
<j`ey> desktop type stuff wont take 5 years
<j`ey> accelerate cyrpto is just normal arm64 stuff
<zid> isn't always normal though
<zid> I'd expect it to work
<j`ey> I mean there's nothing m1 specific there
<zid> nothing said in advance that it had to be though, is my point
<heat> anyway, my OG question: can you notice a slowdown and reduction in battery life or is it totally transparent?
<zid> that's just luck, apple could have added instructions, or used a tpm or whatever
<zid> intel chips have about 839 ways you can fuck up crypto :D
<j`ey> zid: sure, I was just talking about the current hw :P
hypoalgesia has joined #osdev
<j`ey> heat: if you try to run a 4k screen of the mini, or use the imac, apparently you can see a bit of a slowdown
<j`ey> s/of/on
<j`ey> or a bit sluggish, but otherwise, no
<heat> oh yeah that reminds me
<heat> how's scaling for hidpi screens?
<j`ey> sleep2idle doesnt work fully yet, issue in the wifi driver
<heat> no worries s2idle doesn't work in intel either
<j`ey> also screen brightness doesnt work yet, so that doesnt help with battery life
xenos1984 has quit [Ping timeout: 252 seconds]
<zid> j`ey: Just never open the screen, then no need to care about accel or brightness
<j`ey> but the on button is inside
<zid> coathanger
<zid> ssh in, run your nethacks and your zorks
<j`ey> or just never turn it on
<mjg> apparently you can release linux on it
<zid> or never turn it off, it comes with osx I assume
<mjg> so i presume a full screen terminal works
<mjg> re 22:39 < heat> is linux actually usable on the m1?
<zid> who wants unaccelated term with tearing and stuff
<zid> and uncontrolled brightness
Raito_Bezarius has quit [Ping timeout: 268 seconds]
<zid> laptop max brightness is going to be designed for california sun, not indoors in the UK
<j`ey> mjg: i presume linus is using a full desktop
<mjg> it was not a serous comment
<zid> my eyeballs would literally melt out of my skull
<j`ey> zid: you can boot into macos, reduce the brightness, and then reboot into linux, apparently that works :P
<zid> ah okay then that's not too bad
nyah has joined #osdev
<mjg> and people complained aobut rebooting to windows to play games!
<zid> well I don't use laptops so I don't give a shit about changing it
<zid> just that it'd have a bad default if I used it as a desktop
GeDaMo has quit [Quit: There is as yet insufficient data for a meaningful answer.]
<j`ey> my only annoyance is that I got the 8G model, with 8 cores that's kinda annoying
<heat> "not too bad"
<j`ey> when trying to compile C+
<j`ey> +
<mjg> cores? not threads?
<zid> 8G as in gigabytes of ram?
<zid> in 2022?
<j`ey> mjg: 8 cores
<j`ey> zid: yes :(
<mjg> you definitely can't watch porn!
<zid> wat the decade
<heat> i've got a 4 core 8 thread PoC with 32GB of ram
<heat> note that PoC here doesn't mean proof of concept
<zid> 4 core 8 thread with 24, cus I didn't manage to find 32 for less than $10
<zid> I also have a 6/12 but I need to get a new psu :P
<mjg> man i'm on a haswell laptop right now
<mjg> :>
<heat> lucky you
<zid> sandy
<zid> but not craptop
<mjg> 2 core * 2 threads, on top of all the intel shafting...
<heat> j`ey, they're adding the std crate for UEFI in one of EDK2's GSoC
<heat> you left too early :v
<j`ey> heat: im not sure even rust could make me stay :P
<heat> damn you really didn't like that huh?
<zid> Oh are we playing fo4 now
<zid> j`ey did not like that.
<heat> lol
<j`ey> wtf is fo4
<zid> fallout 4
<zid> it had a dumb reputation system
<j`ey> im not a gamer at all
<heat> bruh
<heat> anyway I'm starting to get pissed off at edk2 as well
<heat> and no one pays me :(
<j`ey> someone pays you, just not to do edk2
<zid> You can pay me to not use uefi if you want
<heat> last week's highlight was "I want you to add ext4 support. me: What's the use case? them: the use case is to write to ext4"
<mjg> why not xfs
<heat> erm sorry, ext4 write support
<mjg> use case: avoid writing to ext4
<heat> context: I'm the maintainer of the ext4 driver
<zid> Have you considered adding write support
<heat> everyone seems to understand that it's a bad idea to add write support
<heat> zid, I have
<zid> what could possibly go wrong
<j`ey> heat: wait.. who asked you to write it??
<mjg> you can even add journaling
<mjg> sounds like a weekend porject to me
<heat> j`ey, to write what?
<heat> yeah journaling is easy, just like write support
<heat> you just add it
<heat> done
<zid> It's *vaguely* annoying that grub has no write support, for if you forget to add your kernel to menu.lst and have to type it in to boot it, then type it again once the system boots into nano I guess?
<j`ey> heat: "I want you to add ext4 support." I assume someone was asking you to add it?
<heat> yes, some downstream person
<heat> marvell
<j`ey> heat: wow
<heat> privately
<zid> Did they say what the price was
<zid> for their private patch
<heat> of course not
<heat> I'm an open source maintainer
<heat> I work for free!
<zid> sounds like they don't want it much then
<heat> I also spent some time debugging an issue they had which ended up being because of a storage driver bug.......
Raito_Bezarius has joined #osdev
<heat> although all-in-all most downstream people I know of are nice
<zid> heat: I am a downstream people though
<zid> so you have to weigh it against that
<clever> zid: grub2 does have write support, but its limited, the saveenv/loadenv functions
<clever> zid: it cant change the length of a file, so the file has to be padded out with #'s from within linux
<clever> and then when you saveenv, it just does an in-place write of that sector
<zid> yea that's what I was thnking for write support
<zid> for ext4
<zid> I could do that as long as I didn't have to you know, make things bigger :p
<clever> it can be used for basic things, like saving the default option
<zid> basically nvram
Raito_Bezarius has quit [Max SendQ exceeded]
<clever> but more complex like editing kernel args, it wouldnt really work
Raito_Bezarius has joined #osdev
<heat> yeah sure but that's not really write support
<zid> correct
<zid> It's much more like nvram on a consumer router or whatever
<clever> yeah
<clever> that reminds me, many years ago, i found that i could `modprobe nvram` and then interact with `/dev/nvmram` i think it was, and see bytes that are linked to the bios pw
<clever> but it also had a checksum, so attempting to change the pw i didnt know, caused it to just factory reset itself
<clever> which was just as good
skipwich has quit [Ping timeout: 252 seconds]
<heat> some new AMI firmware does discovery of boot options
<heat> for some reason
<heat> I don't trust downstream vendors
<heat> they are so incompetent it's not even funny
<zid> where's the nvram typically exposed from, superio type thing?
<clever> on the laptop i did it on, i think it was sram inside the RTC
skipwich has joined #osdev
<clever> backed by the same battery that ran the clock
<heat> probably the LPC
<zid> yea that's where it lives, where is it exposed from?
<zid> LPC was my guess
<zid> you can't just steal it like that heat
<heat> i just did
<heat> what now
<zid> The Low Pin Count (LPC) bus is a computer bus used on IBM-compatible personal computers to connect low-bandwidth devices to the CPU, such as the BIOS ROM (BIOS ROM was moved to the Serial Peripheral Interface (SPI) bus in 2006[1]), "legacy" I/O devices (integrated into Super I/O, Embedded Controller or IPMI chip), and Trusted Platform Module (TPM)
<zid> you stole my guess
<heat> The Low Pin Count (LPC) bus is a computer bus used on IBM-compatible personal computers to connect low-bandwidth devices to the CPU, such as the BIOS ROM (BIOS ROM was moved to the Serial Peripheral Interface (SPI) bus in 2006[1]), "legacy" I/O devices (integrated into Super I/O, Embedded Controller or IPMI chip), and Trusted Platform Module (TPM)
<zid> I want it back
<heat> no
<heat> iz mine
<zid> fine, I guess it's connected to the cpu.
<zid> Now my guess contains all guesses
<clever> ive asked before in #coreboot, and whichever chip is handling the spi<->lpc conversion, will query the size of the spi flash, and then adjust the base addr
<clever> so the flash ends on the x86 reset vector
<clever> at least, on the pre-ME/PSP models
<heat> yeah
<heat> well, it's all fixed-size flash
<heat> usually 8 or 16MB
<heat> I think relatively modern intel platforms use 16
<clever> leah from #libreboot is upgrading a lot of laptops to 16mb
<clever> so a linux kernel+initrd can be jammed right into the flash
<zid> mine's 32MB
<clever> the fact that you can upgrade it, means its not fixed, in the translation step
<zid> and it's compressed too lol
wootehfoot has quit [Ping timeout: 240 seconds]
<heat> yeah everything is compressed
<heat> edk2 supports LZMA and Brotli
<clever> -rw-r--r-- 1 clever users 2.0M Sep 20 2018 core2quad.bios
<clever> -rw-r--r-- 1 root root 1.0M Feb 25 2018 Supermicro-X6DH8-bios.img
<zid> dang, want one of mine?
<heat> cringe
<clever> i used flashrom to dump 2 of my systems, and its just tiny :P
<clever> but both of those are legacy based bios
<heat> fucking cringe x2
<zid> Actually I am thinking of the wrong board I think, lame
<clever> i misplaced the rom of my efi system, but it was just a metric ton of PE32 files
<heat> yes
<heat> we've went through this
<zid> 8MB on this board I am a big fat liar
<heat> phony
<heat> a big fat phony
<zid> I think I am running an ancient version because I was having issues getting it to train my pci-e or dram or something and it wouldn't post
<zid> on the latest
<zid> they have a .ROM version and then half way through it switches to some .CAP format, and you have to flash a specific version that understands both to be able to switch, it's weird
<heat> is that chipset pre-FSP?
<zid> front side potato?
<heat> firmware support package
<heat> intel's AGESA
<zid> not a clue
<zid> give me a year
<heat> 1939
<zid> I am post 1939 yes
<zid> 2. BIOS structure changed from .ROM to .CAP for Windows 8 full-functionality.
<heat> but it looks like it was created around 2014 or 2013
<zid> pre then
<heat> aw jeez its a packet capture now?
<zid> apparently
* zid runs strings on his bios
<bslsk05> ​gist.github.com: strings.txt · GitHub
<zid> WuuWuuuuuuuWuWuuuuuuWuuuWWuWuWuWuWuuuWuWuuuuuWWWWWWuuuuuuuuuuuuuuuuuuuuuuuuWuWW8!$bV
<heat> that's uninteresting
<heat> try UTF-16 strings
<zid> oh now how do I switch that
<zid> -eb?
<heat> -el?
<heat> l for little
wootehfoot has joined #osdev
<zid> ah yea
<bslsk05> ​gist.github.com: strings.txt · GitHub
<zid> The only reason I kept it around is that I needed to open it in a hex editeur and make sure my microcode was in it
<zid> I was going to *add* it but it was already there, rather
<zid> I was going to murder support for 2600k and change it to E5-1620 but it already knew about them
<heat> ah now that's a riveting string dump
<zid> I prefer the other one
<zid> ... DEBUG BUFFER OVERFLOW!!!
<clever> ELLI LAGDAOLOTS/E ERPECXNOIT:CP PE
<clever> but how do you deal with BE vs LE strings?
<clever> where each char[4] is byte-swapped!
gmacd has joined #osdev
<zid> I wonder what a Q006.ROM is
<clever> i had written a custom until to byte-swap the entire file, and the above then becomes: ILLEGAL LOAD/STORE EXCEPTION PC:
<zid> I see .text I awnna dump the PE out and see what it is
<zid> appears a fair few times actually
<zid> does MZ/PE have a length offset I can borrow to dump these with, or should I find out how to install binwalk
<clever> zid: in the past i used binwalk, but i assume there is an offset+length for each section, like elf has, i also assume EDK2 has some standardized filesystem
<zid> I dun wanna do each section
<zid> that's effort
<heat> edk2 has a standardized filesystem
<clever> but you can then take the highest offset+length, and assume that is the eof
<heat> I'm fairly sure I've went through this with you clever
<zid> okay it won't install because I did emerge --sync yesterday and it wants to change me from python 3.9 to 3.10 first
<zid> ofc
<heat> uefitool helps
<bslsk05> ​LongSoft/UEFITool - UEFI firmware image viewer and editor (502 forks/2844 stargazers/BSD-2-Clause)
<clever> ah
<zid> Lol, first package it tried to install was rust, which *immediately* failed. I'm guessing I failed the disc space check
<zid> yep, I don't have the required 12GB free
psykose_ has joined #osdev
kazinsal_ has joined #osdev
kaichiuchi_ has joined #osdev
* kingoffrance makes impossible to tell if joking face
mats1_ has joined #osdev
theruran_ has joined #osdev
terrorjack5 has joined #osdev
<zid> mount /dev/sdc1 /var/tmp/portage ..let's try again
DrPatater has quit [Ping timeout: 252 seconds]
night_ has joined #osdev
ornitorrincos_ has joined #osdev
jstoker has quit [Ping timeout: 260 seconds]
aejsmith has quit [Ping timeout: 272 seconds]
hypoalge1 has joined #osdev
Test_User has joined #osdev
gruetze_ has joined #osdev
froggey has quit [Ping timeout: 260 seconds]
<zid> okay so in about 2-3 weeks once rust and python are both updated I might be able to try for binwalk
_koolazer has joined #osdev
<heat> try uefitool
<zid> is it written in rust
<heat> no
<heat> it uses Qt
<zid> oh I can't run that
<heat> gentoo is bad
<zid> I don't have qt or boost
<heat> use arch
<zid> nor do I want it
<heat> arch linux
<zid> I have boost masked, it's a pretty effective filter for terrible software
<heat> architecture linux
<heat> fair
<zid> I thought it was "software that collapses if a single piece fails" linux
<heat> no that's just linux
<zid> no that's just systemd
JTL1 has joined #osdev
<heat> systemd bad
<zid> rust good?
<heat> ah shit
<heat> we should rewrite systemd in rust
<heat> fuck yeah
hypoalgesia has quit [*.net *.split]
psykose has quit [*.net *.split]
terrorjack has quit [*.net *.split]
\Test_User has quit [*.net *.split]
[com]buster has quit [*.net *.split]
night has quit [*.net *.split]
clever has quit [*.net *.split]
theruran has quit [*.net *.split]
koolazer has quit [*.net *.split]
ornitorrincos has quit [*.net *.split]
gruetzkopf has quit [*.net *.split]
bytefire has quit [*.net *.split]
JTL has quit [*.net *.split]
kori has quit [*.net *.split]
kaichiuchi has quit [*.net *.split]
kazinsal has quit [*.net *.split]
mats1 has quit [*.net *.split]
hbag has quit [*.net *.split]
psykose_ is now known as psykose
terrorjack5 is now known as terrorjack
theruran_ is now known as theruran
kazinsal_ is now known as kazinsal
kaichiuchi_ is now known as kaichiuchi
mats1_ is now known as mats1
<zid> I love the use of may here
clever_ has joined #osdev
<clever_> i'm back!
clever_ is now known as clever
<zid> boo- hiii
<clever> # flashrom -p internal --read /media/videos/4tb/bios/nas.rom
<clever> this worked on my nas, and i now have an 8mb file
<heat> a second without clever is a second without a stockholm-syndromed person
<zid> millibit?
<clever> hey, i'm RE'ing x86 stuff today, not broadcom :P
<clever> Found Winbond flash chip "W25Q64.V" (8192 kB, SPI) mapped at physical address 0x00000000ff800000.
<zid> ah MiB!
<zid> man, perl gives a LOT of warnings when it compiles
<zid> terrible language
<zid> language quality == how many warnings you get when you build their compiler with gcc
<heat> how many days does it take for you to compile a web browser?
<clever> heat: i compiled chrome once, and i think it took 48h
<zid> web browsers cannot be compiled by mortals
<zid> you need 128GB of ram to link them
<heat> do you reject the internet then
<clever> zid: i remember one day, when i tried to compile 32bit firefox on gentoo, the bloody linker ran out of ram (3g/1g split)
<zid> It's the only thing I use binary packages for
<heat> or use lynx
<heat> cringe
<zid> not that I often use a web browser on a linux system, I tried to avoid it if possible
<heat> definitely not based
<zid> web browsers (both of them)
<zid> are the single most complex piece of software ever written
<zid> by a large margin
<clever> [nix-shell:/media/videos/4tb/bios/nas]$ UEFIExtract ../nas.rom
<zid> we should be doing browserdev not osdev if we want to write projects we can never hope to finish
<clever> the stupid thing out-smarted me, it ignored the nas/ i made, and dumped to ../nas.rom.dump/ lol
<heat> zid, why not both, like serenityOS
<zid> I remember a tweet or something from them about that
<heat> serenityOS basically turned into a browser engine project
<zid> basically saying that reaching acid3 or whatever was a parlor trick, but making it work on the modern web was basically impossible without an entire engineering firm
<j`ey> heat: heh
<zid> https://drewdevault.com/2020/03/18/Reckless-limitless-scope.html This is a link I remember reading at least
<bslsk05> ​drewdevault.com: The reckless, infinite scope of web browsers
skipwich has quit [Quit: DISCONNECT]
<zid> 114M words just for the specs a browser needs, surpassing C, C++, posix, uefi, usb, and every single RFC document added up significantly
<zid> today
skipwich has joined #osdev
<heat> thats dope
<zid> by the time you got through 10% you'd be behind another 20% no doubt
<heat> that's why I write an OS
<heat> shit moves slowly because people hate change
<zid> web specs add a new posix worth of specs every 4 months lol
socksonme_ has quit [Ping timeout: 268 seconds]
<zid> If web browsers sure ain't a mess, they'll do til the mess gets here
skipwich has quit [Ping timeout: 240 seconds]
skipwich has joined #osdev
gmacd has quit [Quit: Ping timeout (120 seconds)]
wootehfoot has quit [Quit: Leaving]
Test_User is now known as \Test_User
<mrvn> Web browsers have multiple domain specific languages and a whole virtual machine in them. You are writing your own emulator.
<zid> How does kernel decide if use cpu or gpu after fetch decode excecute phase?
<zid> ?? pls answer fast
<mrvn> parse error
<mrvn> best guess for that word salat: It doesn't.
<clever> the cpu and gpu are both running in parallel, doing fetch&execute
<mrvn> each pu has it's own program to run
<clever> yep
<zid> clever how did that manage to bait you
<zid> it's obviously a paste from someone who isn't me, surely
<zid> or am I that dumb
<clever> i'm just bored :P
Vercas6 has joined #osdev
Vercas6 has quit [Remote host closed the connection]
Vercas6 has joined #osdev
hypoalge1 has quit [Ping timeout: 268 seconds]
Patater has joined #osdev
<moon-child> zid please , try to understand
<mats1> don't tell him what to do
<moon-child> the kernel must fetch the instruction to the .gpu. in order to decode it
gildasio has quit [Remote host closed the connection]
Vercas6 has quit [Remote host closed the connection]
gildasio has joined #osdev
Vercas6 has joined #osdev
aejsmith has joined #osdev
dayimproper has joined #osdev
Patater has quit [Remote host closed the connection]
pentium has joined #osdev
<pentium> Hello
Patater has joined #osdev
sprocket has joined #osdev
sprock has quit [Ping timeout: 268 seconds]
foudfou has joined #osdev
jstoker has joined #osdev
foudfou_ has quit [Ping timeout: 268 seconds]
pentium has quit [Quit: Lost terminal]
DrPatater has joined #osdev
Patater has quit [Ping timeout: 252 seconds]
foudfou has quit [Remote host closed the connection]
foudfou has joined #osdev
eck has quit [Quit: PIRCH98:WIN 95/98/WIN NT:1.0 (build 1.0.1.1190)]
eck has joined #osdev