marcan changed the topic of #asahi to: Asahi Linux: porting Linux to Apple Silicon macs | Not ready for end users / self contained install yet. Soon. | General project discussion | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Topics: #asahi-dev #asahi-re #asahi-gpu #asahi-stream #asahi-offtopic | Keep things on topic | Logs: https://alx.sh/l/asahi
jmr2 has joined #asahi
<jmr2>
jeffmiw: are you starting from scratch, or from Corellium's ?
David[m]12345678 has quit [Server closed connection]
David[m]12345678 has joined #asahi
<Glanzmann>
marcan[m]: Do I see it correctly that the SMC driver currently does only provide the interfaces and now the consumers like wifi driver need to use it, to power on its power. Or the watchdog driver to turn off the power of the system?
<Glanzmann>
And I think theere was one guy writing a library so that we can use it to get the battery information?
ogimgd[m] has quit [Server closed connection]
ogimgd[m] has joined #asahi
jbowen has joined #asahi
<chadmed>
Glanzmann: i think his question was more to do with why we need it on these machines, not what the SPMI actually _is_
<chadmed>
i.e. what specifically do we need to talk to over SPMI
<marcan>
since I have modules disabled EXPORT_SYMBOL probably compiles out
<marcan>
Glanzmann: no
<marcan>
the SMC driver is not currently hooked up to pcie
<marcan>
so that needs to happen before it's useful
<landscape15[m]>
@_oftc_joske:matrix.org: do you get that compiling ```Image modules dtbs```? Because for me seems to work without compiling modules
kholtastrophe[m] has quit [Server closed connection]
kholtastrophe[m] has joined #asahi
Bicet[m] has quit [Server closed connection]
Bicet[m] has joined #asahi
alexstore06[m] has quit [Server closed connection]
alexstore06[m] has joined #asahi
wollymilkcap[m] has quit [Server closed connection]
wollymilkcap[m] has joined #asahi
BlazinWeed[m] has quit [Server closed connection]
BlazinWeed[m] has joined #asahi
jbowen has joined #asahi
<landscape15[m]>
Glanzmann: For debugging purpose, I would add to enable CONFIG_USB_ANNOUNCE_NEW_DEVICES in Asahi docs. When I have time I try the latest patches and inform you.
sppdqd[m] has quit [Server closed connection]
sppdqd[m] has joined #asahi
josipknezovic[m] has quit [Server closed connection]
josipknezovic[m] has joined #asahi
jbowen has quit [Ping timeout: 480 seconds]
Major_Biscuit has joined #asahi
<Glanzmann>
landscape15[m]: Add whatever you think is useful. I just started the list because so often people (including me) forgot to set a config option and as a result something was not working.
<landscape15[m]>
ok thanks, I’ll do
<Glanzmann>
chadmed / DmitrySboychakov: IIRC jeffmiw has a userland programm to query different smc values, but maybe that was someone else.
<j`ey>
it was joske, but it's for macOS
NotHere[m] has quit [Server closed connection]
NotHere[m] has joined #asahi
JamieMagee[m] has quit [Server closed connection]
JamieMagee[m] has joined #asahi
M32thSystem[m] has quit [Server closed connection]
M32thSystem[m] has joined #asahi
citruscitrus[m] has quit [Server closed connection]
citruscitrus[m] has joined #asahi
cangurbuz[m] has quit [Server closed connection]
cangurbuz[m] has joined #asahi
elosant[m] has quit [Server closed connection]
elosant[m] has joined #asahi
ytg has quit [Server closed connection]
ytg has joined #asahi
isitaboat[m] has quit [Server closed connection]
isitaboat[m] has joined #asahi
lemonsus[m] has quit [Server closed connection]
lemonsus[m] has joined #asahi
dsrt^ has quit [Remote host closed the connection]
dsrt^ has joined #asahi
rethematrix[m] has quit [Server closed connection]
rethematrix[m] has joined #asahi
long[m] has quit [Server closed connection]
long[m] has joined #asahi
matthewayers[m] has quit [Server closed connection]
matthewayers[m] has joined #asahi
Santiago[m]1 has quit [Server closed connection]
Santiago[m]1 has joined #asahi
yoodee[m] has quit [Server closed connection]
yoodee[m] has joined #asahi
GenieLotion[m] has quit [Server closed connection]
GenieLotion[m] has joined #asahi
ryanhrob[m] has quit [Server closed connection]
ryanhrob[m] has joined #asahi
roxiun[m] has quit [Server closed connection]
roxiun[m] has joined #asahi
dsrt^ has quit [Ping timeout: 480 seconds]
dsrt^ has joined #asahi
essentia[m] has quit [Server closed connection]
essentia[m] has joined #asahi
<Glanzmann>
j`ey: Thank you, I see.
fionera[m] has quit [Server closed connection]
fionera[m] has joined #asahi
nirusu[m] has quit [Server closed connection]
nirusu[m] has joined #asahi
dsrt^ has quit [Ping timeout: 480 seconds]
ghantaz[m] has quit [Server closed connection]
ghantaz[m] has joined #asahi
why2[m] has quit [Server closed connection]
why2[m] has joined #asahi
sproede[m] has quit [Server closed connection]
sproede[m] has joined #asahi
angustrau[m] has quit [Server closed connection]
angustrau[m] has joined #asahi
dek[m] has quit [Server closed connection]
dek[m] has joined #asahi
haterade[m] has quit [Server closed connection]
haterade[m] has joined #asahi
nykma[m] has quit [Server closed connection]
nykma[m] has joined #asahi
Rhys[m] is now known as Rhys[m]1
Rhys[m]1 is now known as Rhys[m]12
cavoirom[m] has quit [Server closed connection]
cavoirom[m] has joined #asahi
Rhys[m]12 has quit [Quit: Reconnecting]
Rhys[m]12 has joined #asahi
gladiac has quit [Quit: Ping timeout (120 seconds)]
gladiac has joined #asahi
jbowen has joined #asahi
mlq_ has quit [Server closed connection]
mlq_ has joined #asahi
jbowen has quit [Ping timeout: 480 seconds]
<chadmed>
SMC isnt connected over SPMI though is it?
<sven>
no
<sven>
it's just yet another co-processor running rtkit
yrlf has quit [Server closed connection]
yrlf has joined #asahi
<Glanzmann>
j`ey: cat /sys/class/power_supply/macsmc-battery/capacity is this exposed by default when you enable smc or did you had to enable something?
<j`ey>
Glanzmann: I just wrote a little driver for it, dunno if I will keep working on it though
<Glanzmann>
j`ey: git-format-patch -1; and upload please, if you don't mind. :-)
<Glanzmann>
s/-//
<Glanzmann>
I'm looking forward to the patch that enables wifi. That gives untethered access to Jens Axboe. ;-)
<j`ey>
Glanzmann: maybe if i work on it a bit more, it was just a quick test for now
<Glanzmann>
I see.
<Glanzmann>
And having Jens running Linux on m1 can't hurt.
<j`ey>
next stop: linus
suricato has quit [Server closed connection]
suricato has joined #asahi
<mps>
SMC is merged?
<mps>
or it is somewhere else?
<j`ey>
just pushed to a branch, but there's no interation, so nothing to test
<mps>
j`ey: which branch?
<j`ey>
smc/work
<mps>
aha I see, thnaks
<mps>
s/thnaks/thanks/
hellsenberg is now known as hell__
JoshuaAs- has quit [Server closed connection]
JoshuaAshton has joined #asahi
coolshaurya has quit [Server closed connection]
coolshaurya has joined #asahi
joske has joined #asahi
<joske>
j`ey: I'm also interested in that 'little driver' ;-)
<joske>
I built smc/work, but I didn't have that file, now I know why ;-)
<mps>
joske: are patches for keyboard needed on top of smc/work
<j`ey>
smc/work is just a few patches on top of the asahi branch, so if you had some of the spi and w/e patchs, theyre still needed
<kettenis>
in its current state smc/work doesn't really add anything useful for end-users
<joske>
mps: yes
<joske>
audio too
<joske>
I noticed that audio took a while to initialize
<joske>
and wifi didn't work
<j`ey>
joske: wifi didnt work.. did you have it working before on asahi/asahi?
<joske>
sure
<joske>
running on wifi now ;)
<joske>
running the asahi branch with the input/audio and 4k patches
<joske>
this is on an air
<nsklaus>
joske: you mentionned audio input, i thought it was only audio out until now, does audio in starts working too ?
<j`ey>
input meaning keyboard
<chadmed>
he means input as in keyboard
<nsklaus>
ah i see
<nsklaus>
it's nice keyboard and trackpad got in, once more step toward a base system beta release of asahi
jato has quit [Server closed connection]
jato has joined #asahi
<chadmed>
kb and touchpad have been in linux-asahi for a little while now actually
<chadmed>
i believe some cleanup is required before its upstreamable though
<nsklaus>
we poor users didn't get to know about it until recently..
* nsklaus
point finger at wiki task list and blog's progress reports lacking a bit behind
<j`ey>
complaining wont make them happen quicker :P
<chadmed>
it was in the last blog post, written in early december
<chadmed>
and the wiki has had it listed as supported since at least i edited tasks and wrote Feature Support
<nsklaus>
i admit i prefer git commits gets in before news articles commenting on them
<_jannau_>
it should have been inthe november blog post. My understanding of the task list that task are done when they are in the upstream kernel
<j`ey>
hopefully spi, input and wifi can be in for 5.18, that would be a decent amount of patches from linux/asahi
<nsklaus>
i'm just a bit sad that once the job got done, there's no hurry to share the joy to the rest of the world
<_jannau_>
at least for linux kernel driver
<j`ey>
nsklaus: but it's not at the point where it's easy to use, so not much point in sharing!
<j`ey>
once it's in mainline is the best point to share
<nsklaus>
but i agree, the main thing is the job getting done, and that is what is being done, so that's good ..
Retr0id has quit [Remote host closed the connection]
Retr0id3 is now known as Retr0id
<chadmed>
the driver being committed to linux-asahi doesnt mean its ready for announcing to the world, or for wide user adoption. we've been testing it and it "works" but it spams dmesg about CRC errors from the trackpad. if it got announced for all to come in and use, we would be flooded with 10 quadrillion github issues about it i guarantee you
<nsklaus>
you understand this article you wrote out of noticing many users out there are asking for news and you feel it's a (minor form) of nuisance. instead of writing such article telling users not to ask, best would be to keep news up to date with commits and general progress then point users to those up to date news article
<nsklaus>
s
<nsklaus>
*articles
<nsklaus>
if twitters post or wiki was up to date, all you would have to do is point everyone to the same link
<nsklaus>
you could even put the link in the channel topic and that would be the end of the story
<chadmed>
i think developer time is better spent developing than writing puff pieces for the amusement of impatient users who cant be bothered reading the wiki. features will be announced and added to Feature Support/Tasks when they are supported. that is your "regular updates"
<nsklaus>
it's because developer write verbose articles
<nsklaus>
it's nice, but most user just wish to see the tasklist items being checked, or now they are being worked on
<nsklaus>
*know
<chadmed>
1) people have better things to do than log each and every commit on twitter, 2) the wiki is up to date because i spent most of the past week ensuring that, 3) there is a link to the wiki in the channel topic
anon2022[m] has quit [Server closed connection]
anon2022[m] has joined #asahi
<nsklaus>
where is smc in the task list ?
<nsklaus>
see my point ?
leah has quit [Server closed connection]
<chadmed>
youve already been told things get checked off on the task list when they are upstreamed. the smc driver has not yet been upstreamed and its not even *useful* for *anything* yet
leah has joined #asahi
leah is now known as Guest1763
<nsklaus>
yes, but that's just because that tasklist item is not reflecting one element of progress, it keeps track of what has been done and cleared, and not what is being worked on
<chadmed>
would you expect, for example, a game developer to announce that a game is "done" or ready for consumption the minute their engine compiles? no, because an engine that compiles with nothing attached to it is not a game, and its utterly useless to gamers
<nsklaus>
you are correct of course
<matthewayers[m]>
nsklaus: The wiki also mentions that it isn’t the best idea to ask about this kind of stuff in IRC because it will likely lead to a headache.
<nsklaus>
that's why i said tasklist is missing keeping track of the 'in progress' stuffs
<matthewayers[m]>
Again, a lot of people are working on a lot of things and time is better spent working on the project itself instead of documenting every bug fix as it happens.
<chadmed>
imo keeping track of "in progress" stuff invites people to test stuff that isnt ready, which leads to them spamming bug reports for stuff developers already know about. especially when here, "in progress" could mean anything from "it works but a race condition on boot means that 30% of the time it wont start" to "well we know what writing 1s to 0xf00 does"
<nsklaus>
i could take care of bugging devs to get a statement of what they are doing and keep tasklist up to date with it, if you wanted, that should keep it off your shoulders
<chadmed>
well no see my point is the devs dont want to be bugged for statements. things are announced and released when theyre ready.
<chadmed>
again, i updated the wiki specifically to avoid these sorts of conversations polluting IRC
<chadmed>
its pretty easy to parse, "dont ask us when things will be done, and here is why"
<nsklaus>
then it's normal ppl ask for news, if you think it's better to leave them in the dark with partial info, for example last progress report is from november, asahi twitter last post is from december, last youtube vid is from +2 weeks ago, etc .. after some time pass by with no fresh news people start to get hungry and wonder what's happening. seems logical and to be expected
<mps>
nsklaus: heh, I read few addresses with i2c tools and saw some bits and bytes and that is some kind of progress but I wouldn't call it 'work in progress'
<j`ey>
nsklaus: yeah but also be patient, 2 weeks isnt that long time in reality!
<nsklaus>
i agree 2 weeks is not that long, and this project is generally doing good, i agree asking for patience in this context is quite ok. people just are getting interested and want linux on M1 for yesterday :)
<chadmed>
i think its important to understand that not telling people about every single successful register poke or gpio pull up is pretty good practice. users typically dont understand the significance of such efforts (and is why we have to write long winded articles explaining them) and assume that no work is being done on ${IMPORTANT_FEATURE}
jbowen has joined #asahi
<nsklaus>
there could be a reasonable middle ground between teeling everyone when a register gets successfully poked at and a more general sense of what is being worked on with the project at current time
<chadmed>
like what looks better to a user? going away for two/three/four weeks and coming back with firefox playing out of the builtin speakers? or bit banging a 1kHz beep to the speaker amp from m1n1?
blasty has quit [Server closed connection]
blasty has joined #asahi
<chadmed>
the reasonable middle ground is what happens now. nifty/cool advancements are announced on twitter (e.g. marcan showing off plasma on the macbook pro) and the wiki is updated when those things are ready for people to test/use. its fine how it is. youre the only person who has complained this much
<nsklaus>
more high level general statement: as in: povik works on audio. so far audio out through jack works on all hardware, he's currently working on speaker out and audio in. work is in progress. update on this statement will be made as soon as there will be a relevant progress achieved. stay tunned.
<chadmed>
that was already announced via the blog post.
<chadmed>
you see what i mean? stuff like that is announced when its ready to be announced. just because you feel like it's actually ready _now_ does not make that so.
<nsklaus>
you see, a one or two line statement per feature being supported, a general, high level report, it would provide a good sense of what is happening. no need for a super verbose review that take days to be made, and wasting too much time
<nsklaus>
yes i know that was announced
flying_sausages has quit [Server closed connection]
<chadmed>
yeah so read Feature Support. every feature is literally one line in a table. cant get more succinct than that
jbowen has quit [Ping timeout: 480 seconds]
<chadmed>
and it provides you a good sense of whats happening because features that are actually at a stage where theyre ready to be announced to the world are listed as supported in some capacity.
<nsklaus>
i just used it as example to give an image of small, time efficient form of report, that let everyone knows what's going on, without wasting dev time too much
flying_sausages has joined #asahi
<j`ey>
the short form is that feature support wiki, the long form is the blog post
<j`ey>
next!
<nsklaus>
marcan reviews blog article are a very nice to read, but they are a bit overkill and it's easy to understand doing them is a pain and wasting his time quite a bit and so since they are like that he's not happy about doing them
<nsklaus>
hence they drag behind
<j`ey>
is that what he said? or just what you think?
<nsklaus>
it's what i think of course
<j`ey>
anyway, you can want what you want, but you'll get what you get!
<nsklaus>
maybe he like doing them. i wouldn't. i'd prefer short 1 or 2 line per feature. quick 15mn to 30mn to do an overall report
viljami[m] has quit [Server closed connection]
viljami[m] has joined #asahi
<nsklaus>
as a dev i'd view creating verbose, lengthy report as a pain
<chadmed>
ok yeah so read the wiki. theres no point in anyone writing any "reports" when theres nothing to actually report
<chadmed>
the fact that we can poke the smc is totally irrelevant to anyone who would actually read such a user-targeted "report"
<chadmed>
when the smc is actually useful in linux, i will update the wiki. its not though.
<chadmed>
"oh yeah every time you wanna check your battery you have to do run a bunch of commands as root, also it doesnt actually do half of the other stuff we need it to do yet!" is not worthy of a progress report. its silly.
<nsklaus>
i have said what i wanted and provided suggestions to give more frequent, less verbose report. you'll do as you please of course. the decision was always yours (all the devs).
<nsklaus>
personaly i'd go through every member of the team once a week, and ask for a general statement (high level, layman terms) of progress and i'd post just that somewhere online (wiki or blog or twitter whatever). that way news report would always be up to date.
kaprests has quit [Server closed connection]
kaprests has joined #asahi
<matthewayers[m]>
Sorry, but that’s a big ask for this kind of project. A lot of people do this for free and it’s taxing enough as it is. Check feature support or subscribe to the blog for updates, infrequent as they may be, because that’s the best way to figure out what’s going on without learning how the sausage is made.
<nsklaus>
i was not saying that if a dev has nothing new to give as a weekly statement of progress they would get punished or anything ;)
<nsklaus>
some statements could be valid for weeks, even a month.. if longer then that would indicate the task could be breaked into smaller bits though, but that's up to the dev itself to decide if he knows what he's doing
<nsklaus>
maybe after 6months if statement never change, then probably time to wonder what's going on with that dev though but that would be about it. i bet most dev would have a little something to say once a week. noone require them to provide "earth shattering revolutionary breakthrough splitting the atom" statement of progress every week..
<sven>
this is not a company, if you want weekly progress reports from me you better pay me damn well for it.
<nsklaus>
sven you don't understand you're already providing those.
<sven>
yes, i do understand.
<nsklaus>
i'm just talking about organizing them through a schedule
<sven>
<sven> this is not a company, if you want weekly progress reports from me you better pay me damn well for it.
<nsklaus>
you're already proving those, and you didn't ask for anything special, or didn't felt offended or such. if anything with my way of doing it, you'd be asked less often and would have to provide those less often
<sven>
no, the point is that i don't want to be asked about progress
<sven>
i'll share whatever whenever i see fit and I won't change that
<j`ey>
can we just move on, we understand your point, but it's unlikely to happen in the way you want, since it's happening in the way te people doin the work want
<nsklaus>
you'll be asked anyway
<nsklaus>
sven: i mean, not by me of course, but by everyone gravitating around this project, users poping in the channel, or on twitter, etc ..
<sven>
see what j`ey said
<j`ey>
sven: so youre saying we need to get your boss interested in asahi, so we can force you to give status updates >:D
<sven>
j`ey: :D
<nsklaus>
people and logic ..
<matthewayers[m]>
I’ve taken 5 courses in the last year about logic in various contexts… math, computer science, linguistics, and digital circuits. I feel like I could give an attorney a run for his/her money given the right context
<j`ey>
if the context is boolean logic, sure :P
<mps>
logic comes from logos and no one actually knows what *is* logos :)
jbowen has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
jbowen has quit [Ping timeout: 480 seconds]
artemist has quit [Quit: artemist]
nirgo has joined #asahi
jbowen has joined #asahi
artemist has joined #asahi
joske has quit [Quit: Leaving]
<mps>
trying to build smc/work and got this `drivers/platform/apple/smc_core.c:135:1: error: expected ';' before 'struct' 135 | struct apple_smc *apple_smc_probe(struct device *dev, const struct apple_smc_backend_ops *ops, void *cookie)`
<j`ey>
mps: build it builtin or add a ; to the end of line 136 and make line 66 export the right symbol
<mps>
j`ey: you mean line `EXPORT_SYMBOL(apple_smc_write);`?
<j`ey>
yeah, should be apple_smc_rw
<mps>
j`ey: thanks
<M1bn3mar[m]>
hey, im new 2 this how could i compile asahi linux?
<mps>
M1bn3mar[m]: git clone, add some patches and build, Glanzmann have some short guide which patches are needed, so not much complicated if you have experience with building kernels
<M1bn3mar[m]>
mps: ok, thanks
<M1bn3mar[m]>
mps: hey didn't find the video, i searched youtube & lbry
<mps>
M1bn3mar[m]: I think Glanzmann have text file on some url but I don't have this url at hand now
<mps>
will look later
<M1bn3mar[m]>
mps: ok, thank u, looking forward I'm really excited about this XD
Major_Biscuit has quit [Ping timeout: 480 seconds]
<mps>
I got this in dmesg output `macsmc-rtkit 23e050000.smc: error (____ptrval____): No SRAM region`
<mps>
and `macsmc-rtkit: probe of 23e050000.smc failed with error 5`
<mps>
does that means that some modules or options are not enabled in kernel config?
<kettenis>
probably means your dtb is not up to date
<mps>
kettenis: so I have to build new m1n1+u-boot?
<kettenis>
as I tried to point out earlier today, there is nothing really on that branch yet for users to test
<mps>
kettenis: I don't expect it working 'out-of-the-box', just wanted to look at status
<kettenis>
compare the smc node in the device tree you're using with what marcan has in that tree
<mps>
t8103.dtsi is different
<kettenis>
of course it is
<mps>
and small change in t8103-jxxx.dtsi
<mps>
kettenis: do you plan to sync these
<kettenis>
I'm tracking the asahi branch
<mps>
aha, ok
<kettenis>
so once the stuff marcan is working on shows up there, I'm going to adjust things
<mps>
thanks for explaining
m6wiq has joined #asahi
malvo has quit [Ping timeout: 480 seconds]
<landscape15[m]>
I have a doubt. After installing a kernel module, do I need to compile the kernel with ```make modules``` or ```make modules_install```?
<mps>
landscape15[m]: if you installed module you don't need to build it or I don't understand question
<landscape15[m]>
mps: I mean together with kernel Image compiling, ```make Image modules dtbs```
<mps>
landscape15[m]: right, this is for build and after that make modules_install
<mps>
landscape15[m]: here is my script which I run to build all in tarball https://tpaste.us/YEOe
<mps>
could be of use
<mps>
this is how to add needed packages before building on alpine linux
<landscape15[m]>
ok thanks, then I made it right. So modules_install should be used only if I’m using external modules I guess
<psykose>
it installs the things tagged with =m in kconfig
<landscape15[m]>
Does this mean also the Broadcom WiFi module?
<mps>
landscape15[m]: no, make modules_install is to install modules in kernel thres
<mps>
s/thres/tree/
<mps>
`modules_install` is section in kernel Makefile
<landscape15[m]>
mps: yeah, in theory every compiled module should be installed by modules_install
<mps>
landscape15[m]: and in practice
<landscape15[m]>
s/should/is/, s/be//
<landscape15[m]>
Also in practice :)
<landscape15[m]>
I was in doubt because I still couldn’t make USB working with my .config. Everything else seems to work fine.
<mps>
if you build new module for already running kernel you need `modprobe $module_name` after make modules_install
<mps>
and `depmod -a` before that probably
<landscape15[m]>
mps: yeah I used it to install my Ethernet dongle driver