ChanServ changed the topic of #asahi-dev to: Asahi Linux: porting Linux to Apple Silicon macs | General development | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Logs: https://alx.sh/l/asahi-dev
yuyichao has quit [Ping timeout: 480 seconds]
<chadmed> povik: how safe do you reckon the amp gain sliders are? id like to play around a bit this afternoon, but i cant afford to blow the machine up right now since its the only computer i have powerful enough to view CTs and MRIs :P
yuyichao has joined #asahi-dev
PaterTemporalis has quit [Ping timeout: 480 seconds]
hizon has joined #asahi-dev
riker77_ has joined #asahi-dev
riker77 has quit [Ping timeout: 480 seconds]
riker77_ is now known as riker77
phiologe has quit [Ping timeout: 480 seconds]
phiologe has joined #asahi-dev
kov has quit [Quit: Coyote finally caught me]
kov has joined #asahi-dev
gladiac has joined #asahi-dev
<bluetail[m]> vote for safe threshhold that cannot be overriden
<bluetail[m]> but idk how that works. Sonarworks for instance processes the audio itself and then bumps up / down volume to avoid clipping
<chadmed> ive submitted a feature request to pipewire for all this. it will fix many many many issues with userspace audio handling in linux on literally thousands of devices
<chadmed> also just for the record turns out i didnt need to play with the amp gain settings, we should definitely leave them at zero and hide them from the user. my problem was just that i am incredibly stupid and was applying the wrong FIRs. i shipped the wrong ones too
<Glanzmann> chadmed: Do you have a link for the feature request for reference?
mps has quit [Ping timeout: 480 seconds]
<chadmed> pushed some changes to asahi-audio, should sound better than it did before now. FIRs are also installed to the directory i proposed in the feature request
timokrgr has quit [Quit: User left the chat]
mps has joined #asahi-dev
timokrgr has joined #asahi-dev
the_lanetly_052 has joined #asahi-dev
<marcan> I'm in a documentation mood today, so I'm going to write/finish some distro and platform docs
<marcan> would appreciate if someone can proofread/fix them up in the wiki while I work on everything else that's left later/tomorrow
<sven> i can do some proof reading this evening
<chadmed> ive got a few hours to kill too
<j`ey> I can prof reed to
MajorBiscuit has joined #asahi-dev
MajorBiscuit has quit []
MajorBiscuit has joined #asahi-dev
ChaosPrincess has joined #asahi-dev
kameks has joined #asahi-dev
<j`ey> marcan: 'Components starting at m1n1 (stage 2) ' instead of after
<marcan> it's a wiki, please just edit it
<j`ey> ok
<marcan> I'm going to work on another one :)
<marcan> hmpf, I should add extracting AVD and XHCI firmware to the first installer "if I have an extra hour" list... otherwise users are going to be stuck with a trip back to the installer (plus some reinstall or at least firmware upgrade feature) to get those to work
<sven> ... just one more thing ;)
<marcan> shouldn't be hard, it's just annoying because I need to de-img4 the kernel and binary grep it
<marcan> sven: are you volunteering to write it? :D
<sven> uh. i, er, need to bring up displayport!
<marcan> right...
<marcan> :p
<marcan> the cop-out version here is to just copy XNU to /boot/efi/asahi, and when the time comes just tell people to run some one-time python script from linux to fix up what's missing
<marcan> I think I'll do that, because I can't prove firmware extraction works properly without, you know, those drivers
<sven> sounds like a plan!
<marcan> oh also, bluetooth firmware is an issue
<marcan> I know where that is, should probably just outright copy it for now and do the same, leave it for a script to fix manually
<Jamie[m]1> whats your plan for finding AVD firmware marcan?
<j`ey> marcan: read, reads good
<Jamie[m]1> the offsets for all the firmwares for different AVD versions can be found, since each one is at a symbol called avdM3 in the binary
<Jamie[m]1> (sorry will move to -re)
<Glanzmann> marcan: Nice read, so you learned from Debian how to do things right. :-)
<Glanzmann> marcan: Btw. Currently the Debian grub package also patches the efi binary, I have not found out when the behaviour is to put the uuid in grub.cfg chainload and when not (probably when the signed shim package is installed)
<marcan> Glanzmann: Debian patching grub without upstreaming that in forever is not exactly how to do things right :p
<marcan> I'm not even patching anything, I'm just bypassing grub-install
<Jamie[m]1> oh i somehow skipped over the message about just copying xnu, that sounds like a good idea haha
<marcan> Jamie[m]1: my plan was to automate offset extraction on the dev side and have SHA-1 hashes of kernels/blobs for validation baked into the installer, but for now: https://github.com/AsahiLinux/asahi-installer/commit/763933fe3ee142b17c35bd883e44b53588614231
<Jamie[m]1> nice hehe
<Glanzmann> marcan: I see, I use the distro infrastructure. But I have not yet identified when it uses the uuid and when not. I assume it has something todo with uninstalling grub-efi-arm64-signed.
<Glanzmann> marcan: I got bitten in the past when I bypassed the distro infrastrucuture (without knowing that I was doing it) for installing grub, than came a grub update and the system was unbootable.
<marcan> this is arch, there is no distro infrastructure
<Glanzmann> Oh, I see.
<marcan> also I'm trying to get some tech docs written so please reserve the diatribe for after release? :p
<Glanzmann> Will do so. I'm here to learn from the best.
<marcan> I don't know who the best is, but I just spent this morning making lasagna for a friend's birthday and I *really* want to get this released before tomorrow is over ;)
<Glanzmann> Wow. Godspeed.
chengsun_ has joined #asahi-dev
chengsun has quit [Ping timeout: 480 seconds]
<marcan> that was the one I was working on a while ago, I split it off from the distro/open OS talk and changed a few bits
darkapex has quit [Ping timeout: 480 seconds]
darkapex has joined #asahi-dev
<j`ey> marcan: LGTM
PaterTemporalis has joined #asahi-dev
hizon has quit [Read error: Connection reset by peer]
ChaosPrincess has quit [Remote host closed the connection]
kameks has quit [Ping timeout: 480 seconds]
<Jamie[m]1> could do with second-level-bullet-points under “NVMe is a GPT volume and contains”
<Jamie[m]1> (for the contents of iSC and containers)
<j`ey> Jamie[m]1: edit it in!
bisko has joined #asahi-dev
<Jamie[m]1> it’s ambiguous enough in the current state that i’m not confident i could edit it correctly
<j`ey> I'll do it
<j`ey> Jamie[m]1: done
<j`ey> (I redid all of the bullet points)
cyrozap has quit [Quit: Client quit]
cyrozap has joined #asahi-dev
<marcan> cool :)
<marcan> someone should also go through the wiki and just delete old/outdated junk...
hizon has joined #asahi-dev
<yrlf> marcan: you had a typo in the first paragraph ("deal time" instead of "real time")
<j`ey> > Puts up a pretty logo
<j`ey> a very important job of m1n1 :D
<j`ey> (not sarcasm, its a cool logo!)
<yrlf> j`ey: one of the most important ones :D
<yrlf> jannau: I didn't even know a pkgrel of 2.1 was possible o.O
yuyichao has quit [Ping timeout: 480 seconds]
yuyichao has joined #asahi-dev
ChaosPrincess has joined #asahi-dev
joske has joined #asahi-dev
<marcan> going to get some dinner
hizon has quit [Ping timeout: 480 seconds]
joske has quit [Remote host closed the connection]
ChaosPrincess has quit [Remote host closed the connection]
ChaosPrincess has joined #asahi-dev
<j`ey> new page LGTM so far
<ar> marcan: explaining what "fuOS" means might be useful
<povik> marcan: two fixes on top of 'asahi' that should make it into the release: https://github.com/povik/linux/commits/asahi-fixes
<povik> i can remind you later at a time of your choosing
<kettenis> it is explained... somewhere
<ar> j`ey: ok, so linkint to that wiki page on first use of fuOS would be useful
<ar> linking*
<Dementor[m]> Did anyone hear from marcan: recently?
<Dementor[m]> I heard there was a big earthquake in Japan and i hope he's OK
<psykose> that's quite far from tokyo
<Dementor[m]> "A strong earthquake jolted Japan's northeast coast on Wednesday, shaking buildings, leaving parts of Tokyo without power and triggering a tsunami warning."
<Dementor[m]> "Tokyo Electric Power Company 9501.T said that around 2 million households were without power and that it was checking the condition of reactors at the Fukushima plant, NHK said."
<Dementor[m]> It's possible everything is alright with marcan I just wanted to verify
MajorBiscuit has quit [Quit: WeeChat 3.4]
the_lanetly_052 has quit [Ping timeout: 480 seconds]
<sven> huh, so there are actually only two mux modes that macOS actually uses: usb3 and that usb4-tunnel mode that also seems to be required for usb2
m6wiq has joined #asahi-dev
<sven> but the kext knows a third mode. It’s just never used
<sven> *required for usb2 when you don’t want to bring up the usb3 phy to be exact
<j`ey> something debug related?
<j`ey> Dementor[m]: he's ok, just heard from him
<Dementor[m]> j`ey: Good to know I'm happy he's fine.
<Dementor[m]> Thanks for letting me know 🙏
jmr2 has joined #asahi-dev
<marcan> sorry, was AFK for a while. I'm okay, everyone I was worried about is too.
<marcan> ah, that was changed, indeed
jmr2 has quit [Quit: Page closed]
<krbtgt> the new articles are very nice
<bluetail[m]> When I am connected to the thunderbolt network bridge, my internet connection doesn’t seem to work
<bluetail[m]> is this a routing issue ?
<bluetail[m]> how can I solve this on Linux Debian ?
<bluetail[m]> (non asahi )
<j`ey> it's not an asahi issue, so either the -offtopic channel or a general linux channel is better
<bluetail[m]> Sry. I solved it with metrics. You are right. I posted I’m the wrong ch. sorry.
<jannau> u-boot scripts to use the correct ESP: https://github.com/AsahiLinux/u-boot/pull/3
<jannau> using boot scripts works for now but we shouldn't use them long term. upstream might replace the distroboot scripts in the near future
<tpw_rules> jannau: i remember discovering where u-boot searches for the ESP. why not change that?
<jannau> better approach would be to just use u-boots efi bootmgr and teach that about the preferred ESP
<jannau> I started with that but it won't be finished before the initial release
<tpw_rules> ah ok
<jannau> it requires changes to u-boot's efi partition code (the part uuid is currently not parsed)
<jannau> the installer needs to seed BOOT* variable into u-boot's non-volatile UEFI var storage on the ESP
<Glanzmann> jannau: In order to test it, I can just pull it, rebuild, install stage 2. Than delete my esp and recreate and it should boot from usb?
<jannau> Glanzmann: you can also test if it boots from USB if you still have a stage1 m1n1 with upper case UUID
<jannau> simply replacing stage 2 m1n1+uboot works
<Glanzmann> jannau: I just tried and it failed on be but the run_distrobootcmd worked. Did I do something worng?
<Glanzmann> This is whay I used to build: https://pbot.rmdir.de/g8ilzANIorWGaOvx7mRpLA
<Glanzmann> s/be/me/
<Glanzmann> jannau: Output of blkid: https://pbot.rmdir.de/uxKxHSeaJPgjFWJ_H-nSvw
<jannau> in u-boot please type: 'fdt addr ${fdtcontroladdr}'
<jannau> tab complete should work in u-boot
<jannau> 'fdt print /chosen'
<jannau> does that print asahi,efi-system-partition = "...?
<Glanzmann> jannau: Nope.
<Glanzmann> jannau: Might it be that my stage1 is to old, but I took it from the asahi installer that marcan build yesterday.
<jannau> you took just the binary? then it probably misses th appended variables
<Glanzmann> Yes, like that: cat m1n1/build/m1n1.bin `find linux/arch/arm64/boot/dts/apple/ -name \*.dtb` <(gzip -c u-boot/u-boot-nodtb.bin) > u-boot.bin
<jannau> that's ok for the 2nd stage
<jannau> has the 1st stage 'chosen.asahi,efi-system-partition=EFI-PARTITION-PARTUUID' as appended variable?
<Glanzmann> jannau: No, I don't think so just the: cat m1n1-rust.bin <(echo 'chainload=<Partition UUID>;m1n1/boot.bin') > object.bin
<Glanzmann> jannau: I was to fast on the first reboot, here is the output: https://tg.st/u/IMG_20220316_213402357.jpg
<jannau> if the 1st stage misses chosen.asahi,efi-... it can't pass it along to the 2nd stage. the 2nd stage would be responsible to add it to the device tree
<Glanzmann> jannau: So, I need to reinstall the first stage like that? cat m1n1-rust.bin <(echo 'chainload=<Partition UUID>;m1n1/boot.bin') <(echo 'chosen.asahi,efi-system-partition=EFI-PARTITION-PARTUUID') > object.bin
<jannau> yes
<Glanzmann> I see, I'll test. brb.
<Glanzmann> jannau: That worked like a charm, thanks. :-) I did not had the choosen thing due to manually installing m1n1 in order to test for joske.
<jannau> m1 max doesn't have enough IO for the mac studio and has a gl3590 usb3 hub
m6wiq has quit []
kgarrington has joined #asahi-dev
kgarrington has quit [Remote host closed the connection]
joske_ has joined #asahi-dev
<joske_> Glanzmann: now that's a real friend, breaking his own setup to help me ;-)
joske_ has quit []
<heli0s[m]> > Mark Zuckerberg is working on bringing NFT technology to Instagram, as the company continues to build out their metaverse: "But over the next several months, the ability to bring some of your NFTs in [will be added, and], hopefully over time be able to mint things within that environment."
mps_ has joined #asahi-dev
mps has quit [Ping timeout: 480 seconds]
ChaosPrincess has quit [Remote host closed the connection]