#
…

2021-10-14

00:11doggkruse has joined #asahi
00:11robinp has joined #asahi
00:21doggkruse has quit [Ping timeout: 480 seconds]
01:53chadmed has joined #asahi
02:23PhilippvK has joined #asahi
02:26phiologe has quit [Ping timeout: 480 seconds]
02:53nskl has joined #asahi
02:58nsklaus_ has quit [Ping timeout: 480 seconds]
03:26marvin24_ has joined #asahi
03:29marvin24 has quit [Ping timeout: 480 seconds]
03:56kov has quit [Quit: Coyote finally caught me]
03:56kov has joined #asahi
04:22everslick_ has quit [Remote host closed the connection]
04:22everslick_ has joined #asahi
06:29marcannobodynada: are you interested in any particular aspect?
06:30marcanit's pretty much pick something that piques your interest and roll with it (e.g. how povik just picked up audio)
06:31marcanI think we have a bunch of kernel newbies around, so that's no big deal :)
06:31marcankernel programming isn't particularly strange as far as systems programming goes, you get used to the style fairly quickly
06:32marcanof course there's also plenty of room for testing, writing docs, etc :)
06:48skali has joined #asahi
07:04skali has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
07:14svenyeah, if you know systems or similar low-level programming it's just a matter of getting used to how the kernel and the development process works
07:16j_eyYou can also already start getting / building the kernel + m1n1
07:17j_ey(if thats also new to you nobodynada!)
07:45povikalyssa: thanks!
07:51povikalthough the commit hashes still dont seem to match
07:51povikno worries, i will match by messages then
07:53nskl has quit [Ping timeout: 480 seconds]
07:58povikdoes linux really need to put a license-identifier on a makefile which is basically a file listing?
07:59povikis this to appease some automatic tree-wide check?
08:06Nspace has joined #asahi
08:15Nspace has quit [Quit: Nspace]
08:39povikworking in linux repo i cant specify just 6 characters of commit hash like i am used to :O
08:40povik16**6 / 1_000_000 = 16.78
08:40povikchecks out
08:42mazpovik: we had to bump up the default to 12 a while ago.
09:12VinDuv has quit [Ping timeout: 480 seconds]
09:24marcan*very* handy
09:24marcannow I can largely just make all the review feedback changes in my working tree, run that and rebase -i, and be done with the next version
09:25povikmarcan: nice find!
09:30nobodynadamarcan: cool, thanks! I don't have any particular area of interest right now, but I suppose I can get it booting, poke around to learn how stuff works, and then pick something to work on
09:31nobodynadait'll probably be a fair bit of work until an M1X even boots, won't it? like it'll at least need new device trees and chicken bits and stuff
09:32VinDuv has joined #asahi
09:37nskl has joined #asahi
09:57marcanyeah, though expect a nice bring-up stream the day mine arrives :)
09:57marcantbh we can probably get it booting to a terminal in a day
09:57marcanunless something really radical changed
09:58marcanmostly just new DTs, adding chickens to m1n1, fixing any hardcoded device issues (e.g. if UART0 moved), and driver changes where incompatible changes happened
10:01chadmedonly real difference for bring up _should_ be more firestorm cores afaik. i sincerely doubt apple will bother changing anything else that does not need to be changed, especially given their knack for maintaining hardware compatibility
10:02marcandepends on whether this is the "M1X" or the "M2", but yeah
10:02marcanassuming it's the "M1X" then it's probably largely compatible
10:02marcanthat said, there's rumors of multi-die stuff which might be... interesting
10:02marcandepending on how that's implemented
10:03marcanmight end up being the first NUMA ARM64 laptop? :p
10:03chadmedyeah im talking just "M1X" here. pretty sure M2 will be blizzard/avalanche replacement for the M1
10:13nobodynadamakes sense! I should have been in bed 4 hours ago so I'm going to sign off for tonight, but thank you and I hope I can be at least somewhat useful :)
10:16nobodynadaI've been astounded by the rapid progresss you all have been making, and this sort of low-level programming is sort of an aspiration of mine so I'm super excited for the opportunity to learn & contribute
10:16nobodynada has quit [Quit: leaving]
10:19svenmarcan: autofixup looks incredibly useful!
10:19povikwoo! dcp driver works swell here
10:19povikfirefox doesn't exactly like the audio driver
10:20povikthere's something wrong with the driver reporting back the playback progress
10:20poviksound works, but video doesn't start
10:24povikinteresting how it works while obviously being broken :)
10:28marcanso apparently there's this *other* performance config framework called "interconnects" intended for memory bandwidth and such
10:28marcanand OPP also interops with that
10:28marcanso I guess *that* is what the memory controller should use?
10:28marcan... but that means I now need memory bandwidth numbers if I don't want to fake it
10:28marcanI guess I'll have to run some m1n1 tests...
10:29marcanalyssa: did you have the DISP0 regs handy somewhere? I want to just kill the framebuffer in m1n1 (to remove it from the bandwidth equation for testing)
10:30marcanI guess I could probably just powergate DISP0, lol
10:32j_eystream? ~_~
10:33kettenisit'd be a black screen ;)
10:33j_eycalming :D
10:37kettenismarcan: seems you have bitten off quite a big chunk with that cpufreq/mmc work
10:38ketteniss/mmc/mcc/
10:51povikfound the culprit, firefox happy now
10:52j_eypovik: something in your code?
10:53povikyeah, would be very surprised if it wasn't :-p
10:53povikcelebrated a bit too soon, still doesn't work perfectly
11:13povikisn't the hypervisor messing with time?
11:14chadmedpovik: yeah today went really slowly for me, guess marcan's messing with some code :P
11:14povikeveryone hop onto the marcan-blaming time
11:15poviki mean, train
11:15chadmedgit blame marcan
11:18povikchadmed: well there's literally a variable called stolen_time in hv code
11:23chadmedpovik: i might be wrong (im really stupid) but i think that is to mark the time the cpu was stolen from the m1n1 guest
11:24povikyeah i think so too
11:24povikanyway sound in firefox works ok outside of hypervisor
11:24chadmedvia pulse/pipewire?
11:24chadmedor straight to alsa?
11:24poviki think this is straight to alsa, let me check
11:25povikthere's pulseaudio running, this is some default debian setup
11:26chadmeditd be through pulse then yeah
11:26chadmedi actually dont think ff is capable of sinking direct to alsa
11:31marcanpovik: ah yes, I would not be at all surprised if the HV breaks the passage of time
11:31marcanI think to do this properly requires paravirtualized shenanigans
11:31marcanmight be worth implementing :)
11:31marcanI didn't particularly care for now, but if we're doing audio it starts mattering...
11:32marcanpovik: you can undefine TIME_ACCOUNTING at the top of hv_exc.c and try
11:32marcanif that works I should just make it runtime configurable
11:32marcanit's important when running macOS because otherwise timeouts happen
11:32marcan(when certain MMIO is hooked)
11:32poviki don't think the hv is that important for debugging audio
11:32poviktriggered only few crashes so far
11:33povikotherwise i can run outside of hv, and if i run in hv, i now know not to pay attention to stuttering
11:34povikso it does not seem important to have that implemented
11:35marcansure, but if undefining that is an easy fix, might as well just make it a flag :)
11:56alyssapovik: glad to see DCP driver is useful :-p
11:57alyssamarcan: don't know the DISP0 reg to kill the framebuffer the right way. powergating is probably easier.
11:58alyssamarcan: otoh I am interested in the relevant DCP stuff
11:58alyssafor DPMS
12:10marcangoing to stream more MCC / cpufreq shenanigans in a bit (#asahi-stream)
12:32marvin24 has joined #asahi
12:35marvin24_ has quit [Ping timeout: 480 seconds]
12:58aleasto has joined #asahi
13:02povikmarcan: turns out sound works alright in hv if all driver bugs are fixed
13:03povikso we can postpone doing anything about TIME_ACCOUNTING
13:04povikalyssa: it is! i can watch seinfeld episodes on my mac mini now :-p
13:40alyssapovik: :-p
13:46jbowen has joined #asahi
13:55yuyichao_ has quit [Ping timeout: 480 seconds]
14:19jmr has joined #asahi
14:20jmr has quit [Remote host closed the connection]
14:20jmr2 has joined #asahi
14:20jmr2 has quit [Remote host closed the connection]
14:21yuyichao_ has joined #asahi
14:55psykose has quit [Remote host closed the connection]
14:55psykose has joined #asahi
15:00marvin24_ has joined #asahi
15:03marvin24 has quit [Ping timeout: 480 seconds]
15:28yuyichao has joined #asahi
15:32yuyichao_ has quit [Ping timeout: 480 seconds]
15:48bps has quit [Ping timeout: 480 seconds]
15:48yuyichao_ has joined #asahi
15:52yuyichao has quit [Ping timeout: 480 seconds]
15:53sorearwould m1x or m2 have aicv2?
16:09svenyour guess is as good as mine
16:31arwe won't know until first devices ship to someone who knows how to check for this stuff
16:34jbowen has quit [Remote host closed the connection]
16:39jbowen has joined #asahi
16:53marcanso apparently that dramcfg thing is almost entirely bullshit, lol
16:53marcanso far I found one geekbench subtest where it shows a ~8% performance gain, and I'm not so confident about it not being just noise
16:53marcanthe overall scores are within 1% or so
16:55j_eyis that part of the mcc stuff youve already done or something else?
17:03marcanyes, the mcc stuff I already did
17:03marcanseems it is mostly useless
17:04marcan(I was going to redo it on the stream today, but I wanted to get bandwidth numbers... spent a ton of time learning that apparently the M1 is *50% slower* doing memory ops if you unroll more than 4 stp/ldp per iteration?!)
17:04marcan(then after that it turned out the bandwidth was the same either way, including on macos)
18:34Nspace has joined #asahi
18:37yuyichao has joined #asahi
18:40yuyichao_ has quit [Ping timeout: 480 seconds]
18:55Nspace has quit [Quit: Nspace]
18:58Nspace has joined #asahi
19:03Nspace has quit []
19:09houlton[m] has joined #asahi
20:00landscape15[m]marcan: Maybe it's a stupid question, but i'm curious to know why you are using Python to develop the asahi-insaller.
20:00j_eylandscape15[m]: why not?
20:01bps has joined #asahi
20:01landscape15[m]Well, I would use C because it's more "friendly" with the hardware and there are useful libraries.
20:02alyssauseful libraries? for C?
20:02j_eyit's an installer!
20:02j_eyit's just twiddling some things, running some commands
20:02psykosewhat does 'friendly with the hardware' mean
20:02psykoseis this an elaborate meme
20:02jbowen has quit [Quit: leaving]
20:03landscape15[m]alyssa: I mean that it wold be easier to implement in a GUI application.
20:04alyssaeasier to implement a GUI application? C?
20:04j_eylandscape15[m]: the installer is more of a pre-installer for a distro
20:05psykosealyssa: i think this is a secret hush hush version of c they don't tell us about
20:05mjg59landscape15[m]: The Fedora GUI installer is written in Python. There's no need to use C for this kind of thing.
20:07alyssahi mjg59
20:08svenC and easy to implement :D
20:08svenmaybe i learned the wrong kind of C
20:08landscape15[m]mjg59: Yes i agree. It's only that i don't like a lot how Python works. I'm more a Java fan for the GUI
20:09svenwe'll happily make a java (or whatever) gui installer the default if someone implements and maintains it
20:10landscape15[m]sven: Yeah but as you said it's more a pre-installer. The good part comes after.
20:10psykosewhat case are you making other than 'i dont like python'
20:11svenon a more serious note, languages are tools. they all suck in one way or another.
20:12svenand for an installer it really doesn't matter what language it's implemented in
20:13landscape15[m]psykose: It's only a prejudice. I usually think at Python as an ML language. I must learn it sometimes :)
20:13psykosedecisions based on prejudices are usually not very good
20:13* tophevich[m]can't get "happily make default" and "a lightweight installer to have jre as a dependency" to compute
20:14sventhe "lightweight" installer needs to download 1gb or so of data anyway. it truly doesn't matter.
20:15j_eysven: RIRR
20:15j_eyerr RIIR
20:15tophevich[m]I thought java is quite "borked" on macOS
20:15nobodynada has joined #asahi
20:16landscape15[m]sven: As i saw, the installer extract the entire IPSW, but it doesn't take a long time. Only my connection is quite slow.
20:16svenit doesn't even download the entire IPSW
20:28Nspace has joined #asahi
20:33kovis java not borked anywhere? especially for GUI? hehe
20:34landscape15[m]I personally like Java GUI. I don't find it so terrible...
20:35landscape15[m]Moreover it works very well with OOP
20:35kovI'll say this though: starting a new project in C in 2021 should be considered a crime against humanity, especially now that we have Rust
20:35kovit doesn't make sense to choose something that lets you shoot your foot so easily at this point
20:37landscape15[m]kov: So why XNU, the Linux kernel and other OS parts are built in C? I thin it's still the more efficient for low level programming.
20:37j_eythose werent started in 2021
20:37kovlandscape15[m], historical reasons, Rust didn't exist at that time
20:37svenoh dear
20:37kovon a slightly less off topic note, is there a recommended repo/branch for testing with an M1 Air?
20:38sveni think alyssa has a branch somewhere with almost all of our not-yet-upstreamed drivers
20:38j_eybut that doesnt contain keyboard for the air
20:39kovI think I looked at alyssa's branch but it lacked the Air dtb and keyboard, yeah
20:39j_eybut those are easy to cherry-pick
20:39landscape15[m]j_ey: Yeah but it would be a pain to make a full transition, wouln't it...
20:39j_eylandscape15[m]: yes, but you started asking about an installer..
20:39_aliceI think for a fancy GUI installer on macOS of all platforms, wouldn't Obj-C + Cocoa, or Swift + SwiftUI make the most sense?
20:39_aliceperhaps with PyObjC bindings for the existing code
20:39kovI was looking for something more pret-a-manger for a quick validation of my setup, then I can do that cherry-picking
20:40j_eykov: well if you want keyboard, my branch has that, but nothing else.. lol
20:40landscape15[m]j_ey: Yeah i'm off topic. I mean for low level programming it's still quite solid.
20:40_aliceI think it'd be pretty neat to have a fancy install flow that pre-stages the distribution installer for the next reboot after step2.sh, kinda like how there were Windows GUI installers for Linux back in the day
20:41kovj_ey, haha well, I guess that one it is then, then I can try to merge with alyssa's
20:41j_eylandscape15[m]: only because of history
20:41_alice(but I think for now, the python installer is fine, given that Python is already available in the base system)
20:41j_eykov: it'll be easier to use alyssa's and merge the keyboard into that
20:41kovj_ey, k, thanks!
20:43sven_alice: yup, having a fancy installer with a nice UI would be a lot of fun :)
20:43kovj_ey, cool, thank you, I'll give this a go over the weekend more likely, will report back if I get to a many-hacks + Air keyboard branch
20:44j_eykov: many-hacks or whatever her branch is has the pinctrl code, so you only need to cherry-pick a few from the very start of my branch
20:54alyssaj_ey: I don't mind SPI patches in my downstream, I just can't test them
20:56alyssaI also don't know if the DCP driver is working on the MacBooks yet for the same reason.
20:56j_eythose corellium ones seem pretty stable in terms of 'no-one is working on them', and then hopefully it can be replaced later on with a cleaner driver
20:57alyssaLet me know what to cherrypick and I'll take them
20:57alyssaor is it just unmodified corellium spi? i can grab those I guess.
20:58j_eyand their SPI driver commit also contains the dts files, so you'll want to modify it to not include that (like https://gitlab.arm.com/linux-arm/jg-open/-/commit/0449db2db86a70cef334d7a070e6425f79b0e48a)
21:02bps has quit [Ping timeout: 480 seconds]
21:05alyssaj_ey: I'd still like some assurance that the DCP driver works on the laptops
21:06j_eymaybe kov can help out this w/end (too?)
22:48Nspace has quit [Quit: Nspace]
22:53aleasto has quit [Quit: Konversation terminated!]
22:56yuyichao has quit [Ping timeout: 480 seconds]
23:52lonjil has joined #asahi
…