marcan changed the topic of #asahi to: Asahi Linux: porting Linux to Apple Silicon macs | "Does XXX work yet?": https://alx.sh/fs | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Topics: #asahi-dev #asahi-re #asahi-gpu #asahi-alt #asahi-stream #asahi-offtopic | Keep things on topic | Logs: https://alx.sh/l/asahi
chadmed_ has joined #asahi
user982492_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
bluetail has quit [Quit: The Lounge - https://thelounge.chat]
bluetail has joined #asahi
Brainium has joined #asahi
user982492 has joined #asahi
nsklaus has quit [Quit: ZZZzzz…]
Brainium has quit [Quit: Konversation terminated!]
Brainium has joined #asahi
bluetail has quit [Quit: The Lounge - https://thelounge.chat]
bluetail has joined #asahi
user982492_ has joined #asahi
user982492 has quit [Ping timeout: 480 seconds]
Guest11959 has quit []
vx has joined #asahi
vx is now known as Guest11997
gabuscus has quit []
alarumbe has quit [Remote host closed the connection]
alarumbe has joined #asahi
gabuscus has joined #asahi
user982492_ has quit [Quit: Textual IRC Client: www.textualapp.com]
Brainium has quit [Quit: Konversation terminated!]
mexpr has joined #asahi
mexpr_ has quit [Read error: Connection reset by peer]
mexpr_ has joined #asahi
mexpr has quit [Ping timeout: 480 seconds]
marvin24_ has joined #asahi
marvin24 has quit [Ping timeout: 480 seconds]
julio7359 has quit [Ping timeout: 480 seconds]
julio7359 has joined #asahi
julio7359 has quit [Ping timeout: 480 seconds]
luna_ has joined #asahi
<luna_> https://linuxunplugged.com/507 Fedora 38 review of LUP from yesterday uploaded now
cylm has joined #asahi
luna_ has quit []
julio7359 has joined #asahi
chadmed_ has quit [Remote host closed the connection]
guillaume_g has joined #asahi
MichaelLong has quit [Ping timeout: 480 seconds]
giskard has joined #asahi
bps has joined #asahi
Guest11997 has quit []
vx has joined #asahi
vx is now known as Guest12015
tinstew has joined #asahi
julio7359 has quit [Ping timeout: 480 seconds]
coderobe has quit [Quit: Ping timeout (120 seconds)]
jmsunseri has joined #asahi
jmsunseri has quit []
nsklaus has joined #asahi
coderobe has joined #asahi
mkurz has joined #asahi
Guest12015 has quit []
vx has joined #asahi
vx is now known as Guest12025
Dkey has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
tinstew has quit [Ping timeout: 480 seconds]
tinstew has joined #asahi
coderobe has quit [Quit: Ping timeout (120 seconds)]
coderobe has joined #asahi
tobhe has joined #asahi
tobhe_ has quit [Ping timeout: 480 seconds]
ten157237743 has joined #asahi
ten1572377437 has joined #asahi
ten157237743 has quit [Read error: Connection reset by peer]
ten1572377437 is now known as ten157237743
tinstew has quit [Ping timeout: 480 seconds]
hailey has quit [Quit: The Lounge - https://thelounge.chat]
hailey has joined #asahi
<nsklaus> marcan: why not do with the speaker support like what you do with the rest of linux, asahi in general, gpu and so on ? that is: release something basic, safe to use, and improve it over time ? that's what the team did and continue doing so for the gpu, for example.
<chadmed> because its not basic or safe to use yet
<nsklaus> ah, if that would be the case that would be understandable
<marcan> I could make a basic/safe version in an hour or two
<marcan> it would suck and have reviews saying "audio works now on asahi but sounds like crap"
<nsklaus> maybe i had the wrong impression that it would be made to work safely, albeit in a reduced mode, low max volume and all, to ensure no damage would happen
<chadmed> ah yeah and thats the other thing, i interpret "basic and safe" as "safe and will give the press no quarter to complain"
<marcan> yes, that could be done, by dropping the volume like 16dB
<nsklaus> *that it could be
<marcan> which is ridiculous
<chadmed> whats the point in shipping that lmfao everyone would just have a whinge and then mute it
<marcan> exactly.
<nsklaus> maybe ridiculous is better than nothing at all ?
<chadmed> nope
<marcan> no, it's not
<marcan> bad PR is better than no PR
<marcan> we don't live in a vacuum, project image matters
<marcan> er, worse than
<j`ey> how loud is 16dB?
<ChaosPrincess> maybe stop letting hn/reddit backseat project development :P
<chadmed> people are looking for anything to complain about even when the work is incredible and genuinely awe inspiring
<nsklaus> i won't fight of course, i was just inquiring a bit and proposing some suggestions
<j`ey> Idont have much of a reference
<chadmed> imagine if we released some shithouse un-DSPed woofer-only monstrosity just so we can say "look wowee it makes sound"
<nsklaus> chadmed: i'm not complaining at all
<chadmed> oh no i dont mean you
<chadmed> i mean people on the orange site, the red site, et
<chadmed> etc
<nsklaus> i see
<nsklaus> the orange site ;)
<chadmed> point is if we released some horrible sounding thing just to go "well at least it works you ungrateful flogs" then thats a poor reflection on us
<chadmed> whereas when we release something thats good and they _still_ complain, that reflects poorly on _them_
<chadmed> which is important, or you end up with a reputation like manjaro
<chadmed> the DSP works and is good (for the 14" machines and mac mini), the kernel driver works reliably, once the safety daemon is production ready then there wont be any blockers
<chadmed> other than my lack of access to a 16" and 13" machine to work on those ;)
<nsklaus> i don't intend to troll, complain, or start flame bait/war or anything like that, but it seemed to me, that's what you folk did with gpu support, and it was somehow well received as for general critic/opinion about it and asahi in general. your 1st released version were just framebuffer, and nobody (worth mentioning) mocked or complained about it, and now step by step you've improved it and it can now run crisis, for example
<chadmed> supposedly the 14" DSP works fine on the 16" machines according to feedback but i really dont trust like that and i would eventually like to do it properly
<chadmed> i mean thats true but with that comes hightened expectations
<nsklaus> release early, release often mantra comes to mind ;)
<chadmed> agile is not really a very good model for this specifically
<psykose> nsklaus: where patches
<chadmed> because inter alia the things we break by moving fast are peoples $3000 laptops
<nsklaus> chadmed: i agree safety first, that is for sure
<nsklaus> maybe it's just me, but i wouldn't mind a low max volume , very basic speaker support, that no support at all
<nsklaus> i mean, knowing it would improve over time
<chadmed> j`ey: that question is kinda backwards since the dB being referenced there is not dB SPL but dB of signal
<nsklaus> *than
<chadmed> decreasing a signal by 3 dB halves its absolute power
<chadmed> decreasing it by 16 dB is... no bueno
<j`ey> chadmed: oh I misread, dropping it *by* 16dB, I thought he said 16Db as if that was the max volume
<chadmed> oh no lol 16 dB SPL would be like whispering in a quiet office
<chadmed> -16 dB gain on the output is half the max volume which in my use is only barely loud enough to listen to in my quiet room with the window open
<nsklaus> beats nothing ;)
<chadmed> eh without the DSP it really doesnt
<chadmed> its not like there arent alternatives
<nsklaus> you mean that without the dsp the sound would have bad quality ?
<chadmed> the quality is so unbelievably shit you will literally just not use the speakers at all lmfao
<chadmed> [22:30] <chadmed> whats the point in shipping that lmfao everyone would just have a whinge and then mute it
<nsklaus> interesting, i didn't know that
<nsklaus> to be fair, my analogy with gpu and thunderbolt/usb3 support still stands
<chadmed> without proper DSP and routing the "speakers" are just two cheap paper microspeakers hooked up to an amp chip
<nsklaus> what you released first was basic and you improved it incrementaly
<chadmed> they sound worse than a cheap pre-japan-being-good hand-me-down shortwave radio from the early 70s i have
<chadmed> thats not even really an exaggeration
<j`ey> nsklaus: it's just a different situation
<chadmed> nsklaus: because like i said at this point the DSP and kernel are pretty much done and the blocker is safety, which is an absolute blocker
<chadmed> at this point with the whole thing like 90% done (for the important machines) why bother releasing something crap
<chadmed> and going "oh dont worry guys this isnt the REAL speaker support we've spent two years on this is just the fake one"
<chadmed> silly
<nsklaus> you should have released early you wouldn't be in that situation then you are now, that if you release basic after waiting long people will have difficulties to understand
<j`ey> lol
<chadmed> also people are generally cognisant of the sheer amount of work that goes into bringing a gpu into passing the GLES conformance tests from literally nothing
<j`ey> nsklaus: please stop telling people how to do stuff
<chadmed> whereas people have a generally idea that "its just speakers bro just make them doof sound out bro"
<chadmed> unfortunately thats not how microspeakers work and without hundreds of thousands of dollars of tools and signing a bunch of NDAs and also high salaries we cant really do it any faster than we are
<psykose> when did the project manager get hired
<chadmed> were there no viable alternatives maybe it would have been higher up on the priority list but as it stands people have a proper working line out backed by quite an excellent DAC and amp, and bluetooth works pretty great too
<nsklaus> j`ey: yes, i'm not doing anyone how he should do what he does, it's not critic and demands, it was just ideas and suggestion, from someone eager to see the project progress and be able to use it on my m2pro laptop. it's casual talking, i do not wish to teach lessons or best practices or anything.. that was just casual talk
<chadmed> so its not really a priority
<nsklaus> *i'm not doing -> i'm not telling how people should do what they do ...
<j`ey> nsklaus: ok, but you have reiterated your point now, and they have explained why it is being done the way it is
<nsklaus> if i had to say something about the project asahi in general, that would be just praise
<chadmed> we're going to end up going in circles
<jannau> "< nsklaus> you should have released early ..."
<chadmed> but yeah like, people would _not_ have understood if we released earlier in a "safe basic" state because it was never really safe and was never really even at a "basic" level that anyone on this planet would deem shippable
<nsklaus> yes, my point was hear, and i thank you folks for the replies, i'll let you do your magic, and be patient about it, no worries
<nsklaus> *heard
<jannau> bye
<ChaosPrincess> well, there is a great method of making features you want appear :P
<nsklaus> i haven't said i'd leave ? ;)
<ChaosPrincess> make them yoursef
<ChaosPrincess> works 100% of the time
<nsklaus> ChaosPrincess: i agree with what you're saying of course.. i was not making any demands of course, that was just questions concerning the general logic
<chadmed> the logic is sound, and if we had released some abominable woofer-only garbage then everyone would have gone "why didnt you just wait this sucks"
<chadmed> the firmware framebuffer worked out okay because the cpus are screaming fast so apart from the terrible battery life you could barely even tell
<nsklaus> some people are never happy, you shouldn't listen too much about those, not everyone's opinion matter in the same way than some other's ..
<marcan> nsklaus: those people end up writing articles and giving a bad name to the project
<nsklaus> there will always be naysayers
<chadmed> yeah there will, so why give them any ammo
<chadmed> "oh haha yeah beta software your fault if you melt your $3000 laptop speakers lol owned"
<nsklaus> don't mind them, drive your own pace, and the things you do will be the proof in itself that you don't have to feel any guilt or shame or anything from those
<chadmed> this is all tangential anyway since people who care enough can already use the speakers, all the development happens out in the open
<chadmed> but as evidenced by a lack of complaining (mostly) no one actually does, and people seem happy to wait until its good and ready
<nsklaus> frankly speaking, asahi project already proved it was capable, and it delivers. naysayers have no more grounds
<nsklaus> yes, i will happily wait too, but please hurry :D
<nsklaus> i've been prisonner of jail-bsd for too long ;)
<nsklaus> i'm just joking a bit
<marcan> you seem to be completely dismissing PR and project image as if only the software matters.
<marcan> that is not how it works
<marcan> half the reason this project is successful is because I played the right PR cards at the beginning to get it credible
<marcan> good PR also attracts developers
<nsklaus> hmm, those are sensible arguments. i admit i tend to ignore those details myself
<chadmed> most people do which is why theyre not in charge of projects and dont influence when we ship stuff
<marcan> and the next step, now that this stuff is actually getting seriously usable, is to get *non-platform* developers on board buying these things
<marcan> because I can't fix everything that is broken or underdeveloped in desktop linux myself
<marcan> so we need more interest from all the upstreams we interact with
<marcan> and that means we should optimize for another nice media cycle once all the things that are coming together lately get done
felix_ has quit [Remote host closed the connection]
cylm_ has joined #asahi
<marcan> that's how you get things like notch support or nice touchbar integration rolling eventually
<marcan> you need more people involved higher up the stack
<chadmed> hell its how we got half the stuff we needed in pipewire too
<marcan> yup
<chadmed> dmitry sharshakov was interested enough to help out with getting stuff like multiple sample rate IRs upstream
<chadmed> which was a massive blocker
cylm has quit [Ping timeout: 480 seconds]
<marcan> I wish I could just focus on the code and spend all day doing kernel coding, but then the project wouldn't have succeeded, because that is not in fact all that matters
<nsklaus> marcan: you have more than just a technical interest in this i see, you also think of general image and public acceptance .. i admit i'm totaly ignorant of those things. but that is good to hear you're concerned even by those aspects too.. i cannot say anything then. all is well. i should just be patient.
jamespmorgan has joined #asahi
<nsklaus> hopefully in one month or two i should be able to use asahi on m2pro laptop
<jn> meanwhile, external speakers and headphones should work, right?
<nsklaus> jn: headphones yes
<nsklaus> headphone, using the "jack" physical port,
<nsklaus> and also sound through bluetooth worked last time i tried
felix_ has joined #asahi
jamespm__ has quit [Ping timeout: 480 seconds]
felix_ has quit [Quit: felix_]
felix_ has joined #asahi
MajorBiscuit has joined #asahi
marvin24 has joined #asahi
<piroko> is there a way to boost the drive on the headphone jack of say, the 14" macbook pro? I have some 80ohm headphones hooked up and they sound really quiet even at 100% volume
marvin24_ has quit [Ping timeout: 480 seconds]
<chadmed> the controls in alsamixer might have what youre looking for
Agua has joined #asahi
<piroko> that just has a single slider for pipewire
<piroko> oh I can switch it
<chadmed> Fn+F6
<chadmed> yup
<chadmed> you want the Jack DAC Gain slider
<piroko> yuuup perfect, thank you!
<chadmed> no worries
<povik> yeah, and if the controls are not enough, there are options on the codec to boost the signal even further, though they are not supported by the kernel driver yet
coderobe has quit [Quit: Ping timeout (120 seconds)]
coderobe has joined #asahi
possiblemeatball has joined #asahi
possiblemeatball has quit [Quit: Quit]
possiblemeatball has joined #asahi
<marcan> ChaosPrincess, povik, jannau: the genpd dependency probe order mess should be fixed now in asahi-wip. I removed the workaround always-on properties.
<j`ey> marcan: got a weird copy/paste thing in the commit
<j`ey> (after the sign off)
<marcan> the last line? I just fixed that
<j`ey> data race, I pulled at the wrong time :P
<marcan> (also that was intended for -dev)
<_jannau_> btw, the branch asahi is still at asahi-6.1-2 instead of the released asahi-6.2-11 (or asahi-6.2-12)
Knedlik has joined #asahi
Knedlik has quit [Remote host closed the connection]
felix_ has quit [Ping timeout: 480 seconds]
<marcan> ah yeah, fixing
<marcan> uhm. it's not letting me push.
<marcan> but I don't have any explicit branch protection for that branch
<marcan> I added one to allow force pushes and still nothing
<marcan> what...
<marcan> wait but it let me push something else?
<marcan> okay, I have no idea what that was about but it worked when pushing from a branch instead of a tag
<_jannau_> was it trying to push the tag object and not the tree referenced in the tag?
<ChaosPrincess> rebased my patches on new asahi-wip, dcp doesnt want to boot now
<ChaosPrincess> [ 1.410376] apple-dcp 231c00000.dcp: error 0000000016f6349f: Failed to boot RTKit: -62
<ChaosPrincess> [ 1.410417] apple-drm soc:display-subsystem: failed to bind 231c00000.dcp (ops dcp_comp_ops): 186501120
<ChaosPrincess> thats a t8103
* _jannau_ blames genpd
Knedlik has joined #asahi
<ChaosPrincess> and it booted again
<ChaosPrincess> please dont tell me its some reboot flakiness :(
<marcan> _jannau_: ah, I bet that was it yeah
Knedlik has quit [Remote host closed the connection]
<marcan> ChaosPrincess: t8103? I removed the PMP always-on, I think there might have been a dependency on that?
<marcan> try adding ps_pmp to the dependencies of ps_disp0_cpu0
<ChaosPrincess> sec, im testing whether restoring always-on to disp0 will help
<marcan> in principle it should make no difference any more, if the genpd stuff I added works
<marcan> but the pmp thing might
<marcan> %s: Couldn't map PMP Scratch registers.
<marcan> %s: Couldn't map PMP External Interrupt registers.
<marcan> bet it's that
<ChaosPrincess> restored always-on to ps_disp0_fe and ps_disp0_cpu0, worx now
<marcan> didn't you say it booted again once
<_jannau_> dcp has a dependency on pmp on t8103
<marcan> it should depend on probe races if it's the pmp thing, since the GPU does depend on that
<ChaosPrincess> marcan: so, its even weirder
<ChaosPrincess> if i break dfr, it boots
<marcan> as I said, probe order
<marcan> it's random
<marcan> just add the dep :p
<ChaosPrincess> what dep to what again?
<marcan> 00:41:55 <@marcan> try adding ps_pmp to the dependencies of ps_disp0_cpu0
<ChaosPrincess> ps_disp0_fe has a dep on that already, no?
<marcan> oh huh, it does
<marcan> then I don't know
<ChaosPrincess> well, it fixed that, at least i got to 3 reboots with screen coming up
MajorBiscuit has quit [Ping timeout: 480 seconds]
<marcan> I just repro'd but here it died halfway through the modeset
<marcan> I suspect it's a power domain, but maybe not PMP
<marcan> either that or a regression
<marcan> ok, it's doing RPM on DCP... what
<marcan> no, that's not it
<ChaosPrincess> k
Zopolis4_ has joined #asahi
<ChaosPrincess> also, i wish sddm would actually figure out how to not delay reboots by 90s
Agua has quit [Ping timeout: 480 seconds]
lanodan has quit [Ping timeout: 480 seconds]
<_jannau_> ChaosPrincess: supposedly fixed in ssdm main, we only need a release
<ChaosPrincess> oh, so i should wait till 2k25, got it
<_jannau_> should only happen if you reboot/shutdown from plasma. if you log out and then reboot/shutdown from sddm it's fine
<ChaosPrincess> yea, i am rebooting from sddm now, except when i forget
lanodan has joined #asahi
<_jannau_> you culd also switch to gdm
julio7359 has joined #asahi
<marcan> ChaosPrincess: this is a j293?
<ChaosPrincess> yes
<marcan> hm.
<ChaosPrincess> running the thing i linked above
cylm has joined #asahi
<marcan> I think there's a problem on j273 but I didn't *think* it affected j293 but I might be wrong
<marcan> sec
cylm_ has quit [Ping timeout: 480 seconds]
abd has joined #asahi
<marcan> I think my genpd fix isn't complete or broke somehow
julio7359 has quit [Ping timeout: 480 seconds]
guillaume_g has quit []
abd has quit [Ping timeout: 480 seconds]
<marcan> also why are we on #asahi?
abd has joined #asahi
lanodan_ has joined #asahi
MichaelLong has joined #asahi
lanodan has quit [Read error: Connection reset by peer]
abd has quit [Read error: Connection reset by peer]
yuka has quit [Remote host closed the connection]
yuka has joined #asahi
abd has joined #asahi
abd has quit [Remote host closed the connection]
___nick___ has joined #asahi
eroux has quit [Ping timeout: 480 seconds]
lanodan has joined #asahi
lanodan_ has quit [Ping timeout: 480 seconds]
lanodan has quit [Read error: Connection reset by peer]
lanodan has joined #asahi
mkurz has quit [Read error: Connection reset by peer]
mkurz has joined #asahi
<ayke_> chadmed: I would actually be curious how the speakers sound with and without DSP. Doesn't need to be anything fancy, it sounds like even a recording from a phone would give a pretty good idea. Sharing a recording like that might also go a long way to show people why DSP is not optional.
<ayke_> Just an idea
<ayke_> (personally I use external speakers so I don't really care about internal speaker support)
MajorBiscuit has joined #asahi
<opticron> ayke_, the issue was that to make them safe so you literally can't destroy them, max volume has to be very limited so they don't get very loud
<opticron> with the DSP stuff, it will be able to prevent overexcursion (physical contact inside the speaker) and overtemp (burning them out) and so can allow higher volumes
<ayke_> Yeah that's my point, let people hear the very quiet low quality sound vs what's possible with DSP.
<opticron> I think that's eventually going to be how it works. If the safety daemon isn't running, it'll drop the volumes to safe levels
MajorBiscuit has quit [Ping timeout: 480 seconds]
___nick___ has quit []
___nick___ has joined #asahi
___nick___ has quit []
___nick___ has joined #asahi
tinstew has joined #asahi
<marcan> correct, though that's volume only. the DSP is a separate thing.
___nick___ has quit [Ping timeout: 480 seconds]
ril has joined #asahi
ril has quit [Quit: leaving]
coderobe has quit [Quit: Ping timeout (120 seconds)]
coderobe has joined #asahi
chipxxx has joined #asahi
jeremiahkorreck has joined #asahi
thelounge60 has quit [Ping timeout: 480 seconds]
Agua has joined #asahi
jeremiahkorreck is now known as korreckj328
thelounge60 has joined #asahi
bluetail has quit [Quit: The Lounge - https://thelounge.chat]
bluetail has joined #asahi
grgy has quit [Remote host closed the connection]
jamespmo_ has joined #asahi
grgy has joined #asahi
salimterryli has joined #asahi
tinstew has quit [Ping timeout: 480 seconds]
SalimTer- has quit [Ping timeout: 480 seconds]
jamespmorgan has quit [Ping timeout: 480 seconds]
stipa has joined #asahi
tiramirasu has joined #asahi
tiramirasu has quit []
bcrumb has joined #asahi
i509vcb has joined #asahi
possiblemeatball has quit [Quit: Quit]
cylm has quit [Ping timeout: 480 seconds]
possiblemeatball has joined #asahi
Zopolis4_ has quit []
marvin24_ has joined #asahi
marvin24 has quit [Ping timeout: 480 seconds]
bcrumb has quit [Quit: WeeChat 3.8]
chipxxx has quit []
chipxxx has joined #asahi
Guest12025 has quit []
vx has joined #asahi
vx is now known as Guest12065
abby has quit [Remote host closed the connection]
abby_ has quit [Remote host closed the connection]
chipxxx has quit []
abby has joined #asahi
possiblemeatball has quit [Quit: Quit]
<chadmed> ayke_: i could demonstrate the quality differences pretty well but the problem with demonstrating overexcursion is that at volumes where youd be able to hear it over a crappy phone recording, it would permanently damage my machine
nsklaus has quit [Quit: disappeared in a cloud of smoke]
nsklaus has joined #asahi
korreckj_ has joined #asahi
Moprius has joined #asahi
Raito_Bezarius has quit [Remote host closed the connection]
Raito_Bezarius has joined #asahi
Moprius has quit [Quit: bye]
korreckj_ has quit []
<korreckj328> Not sure if this is the right place to ask, as I'm new to IRC. But I've seen posts on social media about the m1 gpus mentioning gaming with fex. Is this the fex that people are mentioning? Just doing some reading. https://github.com/FEX-Emu/FEX
<chadmed> yup
<korreckj328> Thanks
Agua has quit [Remote host closed the connection]