marcan changed the topic of #asahi to: Asahi Linux: porting Linux to Apple Silicon macs | 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
bgb_ has joined #asahi
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
alexstore06 has quit [Ping timeout: 480 seconds]
phire has quit [Ping timeout: 480 seconds]
alexstore06 has joined #asahi
TheLink has quit [Ping timeout: 480 seconds]
TheLink has joined #asahi
alexstore06 has quit [Ping timeout: 480 seconds]
alexstore06 has joined #asahi
user982492 has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
bgb_ has quit [Ping timeout: 480 seconds]
bgb_ has joined #asahi
rohin2 has joined #asahi
<kode54>
so glad I updated to 12.0.1 to find my GPU problem is not fixed
<kode54>
games perform worse, and within 10-15 minutes of playing them, the GPU locks up
<chadmed>
>apple software fixing things
<kode54>
so glad they don't actually expect people to use their software
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
DarkShadow44 has joined #asahi
DarkShadow4444 has quit [Read error: Connection reset by peer]
user982492 has joined #asahi
XeR has quit [Remote host closed the connection]
XeR has joined #asahi
milek7 has quit [Remote host closed the connection]
milek7 has joined #asahi
PhilippvK has joined #asahi
user982492 has quit [Read error: Connection reset by peer]
user982492_ has joined #asahi
phiologe has quit [Ping timeout: 480 seconds]
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi
<citizen1[m]>
given how many linux users choose macbook as their machine in the corporate works (even trovalds does it) i am surprised macan didn't reach his patreon/opencollective goal to develop linux for m1 full time
<chadmed>
its sad but im not surprised tbh. people are slow to part with their cash even in good times. globally, we are currently _not_ in good times.
<chadmed>
cost of living has exploded in recent months too for a lot of folks
<citizen1[m]>
iphone sales disagree
<citizen1[m]>
people spending crazy in that
<chadmed>
the death throes of neoliberalism. it will drag us all down with it
<citizen1[m]>
plus i believe power in numbers. $5 x 100k linux users = $500k
<chadmed>
dont forget a lot of people in many markets get their phones on contract still. even in australia where that isnt common, many people are on rolling 2 year handset replacement contracts where you just get a new phone every two years regardless
<citizen1[m]>
people are just stingy and want everything absolutely free. when they donate a dollar is like they donate their fortunes
<chadmed>
so looking at the sales for those things probably isnt a good indicator of economic health. i prefer looking at metrics like cost of housing as a function of median yearly income and inflation (but real inflation figures not this crap where they just take out components of the index that make it look worse than they want)
<chadmed>
since this time last year the price of bread has more than doubled, used cars are up like 80%, energy prices are up like 40%
<chadmed>
things are rough
<citizen1[m]>
but how can prices increases when people are getting poorer, demand and supply should make prices lower
<citizen1[m]>
if a car used to be $20k in good times, logically who will pay for it at $35 in rough times🤷🏻♂️
<citizen1[m]>
tbf the asahi project needs campaigning. I think many do not know about it or think its dead
<chadmed>
citizen1[m]: consider that maybe the dogma about supply and demand is utter bs :)
<chadmed>
this is getting really off topic but like, economics are a bit more complex than the supply and demand graph. using the used car example, especially in the US and Australia you have enormous captive markets of people who absolutely _must_ own a car to get basically anywhere. so, you as a seller can charge whatever you want and peopel will pay it
<chadmed>
especially if global supply chains for new products are in tatters which means they cant get reasonably priced new cars
yuyichao has joined #asahi
<chadmed>
doubly especially if reserve banks in every developed country are being forced by their governments to print trillions of dollars in cash in the absence of proper fiscal policy, triply especially if those governments also demand that interest rates remain ultra low to encourage people to borrow money and keep cash flowing through their economies
marvin24 has joined #asahi
marvin24_ has quit [Ping timeout: 480 seconds]
kenjigashu has joined #asahi
kenjigashu has quit []
kenjigashu has joined #asahi
kenjigashu has quit []
balrog has quit [Remote host closed the connection]
<nsklaus_>
povik: since you're around, would you mind if i asked how much do you suspect implementing support for the headphone port will be hard thing to do ? if you have time to talk a bit prospective and perceived difficulty about next steps of this sound driver you're making.. if you don't have time, then ignore my question
<povik>
nsklaus_: sure, i have the time, am happy to answer
<povik>
it snhouldn't be too difficult
<povik>
i have a m1n1 python script which outputs audio on the headphone port
<povik>
nothing stops me from doing the same mmio pokes in the linux driver right now and get a driver for the headphone port
<povik>
but i am trying to understand what the mmio pokes do at the moment
<povik>
so that's why i don't jump onto the linux driver just yet
c10l has joined #asahi
c10l has quit []
c10l has joined #asahi
gabuscus has quit [Ping timeout: 480 seconds]
<j_ey>
povik: oh cool, didnt know you got the headphones working
<povik>
nsklaus_: ^^ essentialy i am trying to figure out where do these numbers come from
<povik>
so that the driver is a bit more universal
<nsklaus_>
i see (well, kind of, since i'm not hardware dev, but i see what you mean). so you're doing some form of reverse enginering, chip analysis, to develop this driver ? it's a complete new driver for linux with no previous support ?
<nsklaus_>
you're using marcan's hypervisor and other tools, to analyze the sound chip, what registers it toggle, what info it treats, and how ? that kind of things ?
<povik>
well you have two parts of the problem
<povik>
the SoC part, where you have to convince a peripheral to take the samples from memory and put them onto a serial digital interface
<povik>
and then the codec part, where another IC part on the circuit board takes those samples from the digital interace and creates an analog signal from them
<povik>
in the second part there is for example already a driver for a similar speaker-amp/codec IC in linux
<povik>
so that i just invoked with some small fix to get the speaker working
<povik>
in the first, SoC part, you are starting from scratch
<nsklaus_>
interesting, do you have to fiddle with interupts and ipc stuffs for the various socs elements to talk to each other in the correct time frame and all ?
<povik>
not that much, you have a DMA engine which takes the samples from memory which triggers an interrupt to ask for more samples
<povik>
other than that, you just have to turn on the elements in the correct order
<nsklaus_>
many thanks for the infos, it's interesting to hear those details
<povik>
happy to answer :)
gabuscus has joined #asahi
<nsklaus_>
would be nice to hear from those adventurous enough to already have asahi up and running if they can manage to get headphone port to work, especialy on laptops
<nsklaus_>
although, i'm wondering how much that headphone support, being only in the m1n1 part, also allow it to be used when everything else (linux itself) is fully up and running ?
<nsklaus_>
if i'm understanding correctly, it would just be a matter of running the python script through a m1n1 remote term, even if linux is fully up and running on the target machine ?
<povik>
the python script is just a demo
<nsklaus_>
yes i understand that, but it's a nice test
<povik>
the knowledge there contained needs to be transferred into the linux driver before it's of any use to you on linux
<povik>
so no, it's not just a matter of running the pythons script while linux is booted up :-p
<nsklaus_>
just testing if audio does or does not come out of headphone plug on laptop, even though if right now it's just for testing a sample audio
<povik>
maybe if you had two mac minis!
<povik>
one running linux, the other the m1n1 proxy
<povik>
than you could use the second as a fancy external sound card for the first
<povik>
*then
<nsklaus_>
:)
<nsklaus_>
a 700$ external soundcard
<povik>
well that's how i was using my mini once i got the speaker up initially
<povik>
a form of celebration i suppose
<nsklaus_>
of course yes
<nsklaus_>
i'm disgressing but the silent explosion of joy, when you're a dev, working on something and you finaly manage to get some success after fiddling long hours, you're alone in the night, and it start working, and you feel like fireworks and celebration are in order but noone's around, everyone is sleeping and you must stay quiet, that part alway amused me :)
nuh^ has joined #asahi
gabuscus_ has joined #asahi
<nsklaus_>
but maybe that's just me because i only vaguely know what i'm doing and when by accident i finaly make it work i'm then more joyful than i should be..
gabuscus has quit [Ping timeout: 480 seconds]
<nsklaus_>
anyway, i'm very much looking forward having sound support in asahi, wish you courage and perseverence
gabuscus has joined #asahi
<povik>
nsklaus_: i think we all get joyful when we get things working in the middle of a night :)
gabuscus_ has quit [Ping timeout: 480 seconds]
<povik>
thanks. i will let you know once i get headphones working on the '14 macbook pro i am expecting to arrive
<povik>
otherwise i only have the mini to test on
<nsklaus_>
ah you took one of these new beasts
<nsklaus_>
good to hear
<rkt>
povik: Do you think is there any new model Mac mini this year with M1 max or pro soc?
gabuscus_ has joined #asahi
<povik>
i don't think so. but my guess is as good as anyone elses
<rkt>
povik: Apple release new M1 chip too fast. we can not catch up.
<rkt>
It seems Apple are planning to release M2 chip as fast as macOS and iOS
<rkt>
typo: M1
<povik>
not necessarily we can't catch up. we will see how this goes but i expect most of the work on m1 will transfer to the new chips well
gabuscus has quit [Ping timeout: 480 seconds]
<nsklaus_>
rkt: on the good side, if i'm not mistaken, apple keeps api and registers names and all the same across the same arch. i think marcan talked about that a little while ago ?
<nsklaus_>
so everything should stay more or less quite similar from m1 to m1pro and m1max
<nsklaus_>
..hopefully..
<rkt>
nsklaus_: yeah, i think they will keep instructions set the same for a long time.
<rkt>
But if not, that’s normal, since they just need to update their compiler. :p
gabuscus_ has quit [Remote host closed the connection]
gabuscus has joined #asahi
gabuscus has quit [Read error: Connection reset by peer]
gabuscus has joined #asahi
<nilsi[m]>
nsklaus_
nuh^ has quit [Remote host closed the connection]
alexstor_ has joined #asahi
<chadmed>
rkt: its not so much the instruction set, linux has had arm support for a very very long time. the hard work is platform-specific drivers for things like the GPU, SPI controller, USB host controllers, etc
<chadmed>
the things that actually make a computer useful
<chadmed>
if apple make material changes to these things then these changes have to be reverse engineered and drivers modified to reflect them
alexstore06 has quit [Ping timeout: 480 seconds]
gabuscus_ has joined #asahi
gabuscus_ has quit [Read error: Connection reset by peer]
gabuscus has quit [Ping timeout: 480 seconds]
gabuscus has joined #asahi
aleasto has joined #asahi
<marcan>
rkt: I expect to "catch up" on monday
<marcan>
people vastly overestimate the delta effort required to support newer chips
<marcan>
especially "we just doubled all the things" type changes like this one
<rkt>
marcan: I just read a news from the interenet on Arashi Linux a few days ago, they qouted that you were ordering some m1pro m1max Macs.
<marcan>
the pro is arriving on friday, and I'm going to try to bring it up on monday
<marcan>
it's likely to go on until well into the morning in EU ;)
<marcan>
I have ~8h allocated that day
<j_ey>
lucky monday is my day off!
<marcan>
ha :D
<TheLink>
Bank holiday in my part of Germany
hspak6 has joined #asahi
hspak has quit [Remote host closed the connection]
hspak6 is now known as hspak
hspak has quit [Ping timeout: 480 seconds]
jmr2 has joined #asahi
<jmr2>
agraf: off-topic... I was looking at the status of your patches for the M1 in QEMU. If I understand correctly, everything is merged upstream, just not included in any official release yet?
<agraf>
jmr2: yeah, it will be in the next release. Also you can't boot Windows :)
<jmr2>
Not that it bothers me much, but why is that?
<agraf>
jmr2: (well, you can, with a trick: Use -M virt,virtualization=on. That will switch internal QEMU logic into treating SMC calls as PSCI calls)
<jmr2>
Another patch coming ?
<agraf>
jmr2: Windows is running an SMC call on boot. The QEMU arm maintainer considers the default virt machine to have "no EL3 implemented" which means the SMC instruction is supposed to trigger a #UDEF according to the spec
<agraf>
It's one of those bits in the spec where based on what functionality your CPU has, the hypervisor's ability to trap it changes
yuyichao has quit [Ping timeout: 480 seconds]
<jmr2>
Thanks for the clarifications - and for your persistence in getting those upstream. The number of revisions was impressive.
jmr2 has quit [Remote host closed the connection]
yuyichao has joined #asahi
bps has quit [Ping timeout: 480 seconds]
X-Scale` has joined #asahi
user982492 has joined #asahi
X-Scale has quit [Ping timeout: 480 seconds]
bps has joined #asahi
* kov
sees the discussion on qemu and checks if his patches for keyboard input quality of life have made it yet
___nick___ has joined #asahi
<kov>
nope, I guess the cocoa mantainers are still MIA xD
___nick___ has quit []
___nick___ has joined #asahi
___nick___ has quit []
___nick___ has joined #asahi
bps has quit [Ping timeout: 480 seconds]
nobodyna1a has joined #asahi
nobodynada has quit [Ping timeout: 480 seconds]
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
bgb_ has quit [Ping timeout: 480 seconds]
user982492 has joined #asahi
jacoxon has joined #asahi
jacoxon has quit []
jacoxon has joined #asahi
Nspace has joined #asahi
<tophevich[m]>
TheLink: don't forget that at the time of the stream you will already have changed to CET (the day before will still being on CEST). Meaning on Sunday Japan 11:00 would be 04:00 Germany but on Monday it is 03:00.
<TheLink>
oh my
<TheLink>
given that until earlier today I had totally forgotten about this holiday, this is getting quite complicated :P
<tophevich[m]>
you can't just forget the catholic "day of the dead", a day of rememberance ... (just yanking your chain :D)
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<TheLink>
my grandma wanted me to become a priest, good thing I didn't
<tophevich[m]>
but in all honesty, it is the one "religious" day I can get behind, as it is a free day to remember the dead, which is a good thing independent of religion, it's just that they (religions) monopolized "grieving culture", as large non theist communities don't exist long enough in big enough numbers to develop their own to be generally known or available.
* tophevich[m]
is sorry for going so far off-topic
<tophevich[m]>
(talking about "western" world, as I'm quite ignorant)
jacoxon has quit []
maor26 has quit [Quit: Leaving]
user982492 has joined #asahi
X-Scale has joined #asahi
X-Scale` has quit [Ping timeout: 480 seconds]
jeffmiw has joined #asahi
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
___nick___ has quit [Ping timeout: 480 seconds]
user982492 has joined #asahi
<jeffmiw>
did anyone booted linux under the hv recently ? I must be doing something wrong.
<jeffmiw>
I'm creating my macho this way: cat ../../m1n1_marcan/build/m1n1.macho arch/arm64/boot/Image.gz arch/arm64/boot/dts/apple/t8103-j274.dtb ../../initrd/init-new.gz > m1n1-payload.macho
<jeffmiw>
and doing a run_guest this way: python3.9 ../../m1n1_marcan/proxyclient/tools/run_guest.py m1n1-payload.macho
<j_ey>
jeffmiw: and?
<jeffmiw>
just after jumping to the kernel it is like it is rebooting in m1n1 and stay there
<j_ey>
weird.. did you update m1n1? Im on the latests and its fine
<jeffmiw>
yes I'm chainloading the latest version 0d0a646
<j_ey>
ohhh
<jeffmiw>
ok, if you have it working with similar commands, I'll try to understand why it is not working on my side
<jeffmiw>
same result, I think there is something wrong with my macho payload, m1n1 says it found kernel/devicetree/initramfs but then No valid payload found
<jeffmiw>
yes, you need to put the kernel Image.gz at the end ... so the macho image shall be constructed like this: cat ../../m1n1_marcan/build/m1n1.macho arch/arm64/boot/dts/apple/t8103-j274.dtb ../../initrd/init-new.gz arch/arm64/boot/Image.gz > m1n1-payload.macho
<j_ey>
my image.gz is first
<jeffmiw>
humm first but still after the m1n1.macho ?
Nspace has quit [Quit: Nspace]
<jeffmiw>
I have the linux kernel booting but it is taking over the USB and I'm loosing the HV proxy interface :(
<j_ey>
yeah sorry
<j_ey>
you need to disable the usb ports in the dtb
<j_ey>
or just one of them
<jeffmiw>
so I should turn one of the USB_DWC3/2 off or both to keep one for the hv ?
<j_ey>
i cant remember which port is which, but comment one of them out.. if it doesnt work, its the other one :p
<jeffmiw>
strange, I'm back to the no valid payload found. It seems I need to dig into this one
<jeffmiw>
ok ok ... I needed your patch in payload_run for the m1n1 in the concatenated macho image ... not just chainloading it. now things are making sense
jeffmiw has quit [Remote host closed the connection]
jeffmiw has joined #asahi
<j_ey>
i always chainload and payload the same m1n1
yuyichao has quit [Ping timeout: 480 seconds]
bps has quit [Ping timeout: 480 seconds]
Nspace has joined #asahi
bps has joined #asahi
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
user982492 has joined #asahi
aleasto has quit [Quit: Konversation terminated!]
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]