ChanServ changed the topic of #asahi-dev to: Asahi Linux: porting Linux to Apple Silicon macs | Non-development talk: #asahi | General development | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Logs: https://alx.sh/l/asahi-dev
<tpw_rules>
it erases your hard disk and you have to DFU restore from another machine
<tpw_rules>
and you lose all your data
<kallisti>
Hmm yeah, if I was smart perhaps I could make a backup and then risk it.
<tpw_rules>
i mean i suspect it just won't work. you'd need a device tree and things
<kallisti>
Ahh right, I dumped one of those before, but never made next steps (for a tablet)
<kallisti>
Not sure how one would be created from scratch. That was android.
kallisti has quit [Remote host closed the connection]
gabuscus has joined #asahi-dev
lsep has quit [Ping timeout: 480 seconds]
<nicolas17>
I don't understand, isn't 14" MBP already in that list?
tristan2_ has joined #asahi-dev
cylm_ has joined #asahi-dev
midou has quit [Ping timeout: 480 seconds]
tristan2 has quit [Ping timeout: 480 seconds]
<chadmed>
yes
cylm_ has quit [Quit: WeeChat 4.0.4]
crabbedhaloablut has joined #asahi-dev
loki_val has joined #asahi-dev
crabbedhaloablut has quit [Read error: Connection reset by peer]
roxfan has quit [Ping timeout: 480 seconds]
nicolas17 has quit [Quit: Konversation terminated!]
roxfan has joined #asahi-dev
roxfan has quit [Read error: Connection reset by peer]
qyliss has quit [Quit: bye]
qyliss has joined #asahi-dev
qyliss has quit []
qyliss has joined #asahi-dev
zensayyy has joined #asahi-dev
kali_ has joined #asahi-dev
kali_ has quit []
zensayyy has quit [Quit: zensayyy]
jacksonchen666 has quit [Ping timeout: 480 seconds]
ydalton has joined #asahi-dev
hightower2 has quit [Ping timeout: 480 seconds]
<ydalton>
i assume the only thing you can do regarding linux is to install m1n1, but that's not very useful to an end user
ydalton has left #asahi-dev [ERC 5.4 (IRC client for GNU Emacs 28.2)]
midou has joined #asahi-dev
hightower2 has joined #asahi-dev
eiln has joined #asahi-dev
hightower3 has joined #asahi-dev
hightower2 has quit [Ping timeout: 480 seconds]
eldriitch has joined #asahi-dev
hightower2 has joined #asahi-dev
midou has quit [Ping timeout: 480 seconds]
hightower3 has quit [Ping timeout: 480 seconds]
eiln has quit [Quit: WeeChat 4.0.4]
jacksonchen666 has joined #asahi-dev
SalimTerryLi has quit [Ping timeout: 480 seconds]
<chadmed>
huh? no those laptops are supported fully afaik
<chadmed>
the reason the M2* _desktops_ arent is because of the DCP/dptext/dcpext/whatever mess
roxfan has joined #asahi-dev
hightower2 has quit [Ping timeout: 480 seconds]
<jannau>
bcm4388 is not supported, asahi alarm images are too old for 13.5 OS FW or not yet marked as 13.5 compatible, devices are expert mode only. at least on the m2 pro/fedora there's a strange issue with gpu's uat_handoff reserved memory
<jannau>
so the devices work to some extend but are not officially supported
compassion1785 has quit [Quit: lounge quit]
compassion1785 has joined #asahi-dev
compassion1785 has quit []
compassion1785 has joined #asahi-dev
maria has quit [Ping timeout: 480 seconds]
maria has joined #asahi-dev
roxfan has quit [Remote host closed the connection]
CoolStar has quit [Read error: Connection reset by peer]
SalimTerryLi has joined #asahi-dev
roxfan has joined #asahi-dev
roxfan has quit [Read error: Connection reset by peer]
roxfan has joined #asahi-dev
roxfan2 has joined #asahi-dev
roxfan has quit [Ping timeout: 480 seconds]
roxfan has joined #asahi-dev
roxfan2 has quit [Ping timeout: 480 seconds]
roxfan has quit [Ping timeout: 480 seconds]
aleasto- is now known as aleasto
jeisom has quit [Ping timeout: 480 seconds]
roxfan has joined #asahi-dev
nickchan has quit [Quit: ZNC 1.8.2+deb3.1 - https://znc.in]
nickchan has joined #asahi-dev
roxfan has quit [Ping timeout: 480 seconds]
<jannau>
maz: would you take a simple openscad based tray "case" for the cs-hw repo?
c10l has quit [Quit: Bye o/]
c10l has joined #asahi-dev
c10l has quit []
c10l has joined #asahi-dev
hightower2 has joined #asahi-dev
rhysmdnz has quit [Quit: Bridge terminating on SIGTERM]
Guest250 has quit [Quit: Bridge terminating on SIGTERM]
Jamie has joined #asahi-dev
Jamie is now known as Guest328
rhysmdnz has joined #asahi-dev
lewurm has quit [Read error: No route to host]
lewurm has joined #asahi-dev
eldriitch has quit [Quit: Page closed]
eldriitch has joined #asahi-dev
ellyq has joined #asahi-dev
eldriitch2 has joined #asahi-dev
eldriitch2 has quit [Ping timeout: 480 seconds]
chadmed has quit [Read error: Connection reset by peer]
chadmed has joined #asahi-dev
nicolas17 has joined #asahi-dev
chadmed has quit [Ping timeout: 480 seconds]
roxfan has joined #asahi-dev
roxfan2 has joined #asahi-dev
roxfan has quit [Ping timeout: 480 seconds]
eldriitch has quit [Remote host closed the connection]
<eiln>
now to RE those ~200 magic registers (they're actually dma'd *to* the registers from iova space).. per codec..
<opticron>
congrats! great work!
compassion1785 has quit [Quit: lounge quit]
elros has quit [Ping timeout: 480 seconds]
compassion1785 has joined #asahi-dev
qyliss has quit [Quit: bye]
qyliss has joined #asahi-dev
loki_val has quit []
elros has joined #asahi-dev
compassion1785 has quit [Quit: lounge quit]
jeisom has joined #asahi-dev
<iaguis>
awesome!
elros has quit [Ping timeout: 480 seconds]
<amarioguy>
btw what is "rc" supposed to be in the apple devicetrees? is that just a bespoke mechanism of apple's pcie thing or is that actually in the standard?
<amarioguy>
(for the pcie node)
<amarioguy>
and i'm almost scared to ask...but if apple hw is ecam compliant, a bit curious why we needed to roll our own passthrough driver? just DART/MSI stuff ig?
<amarioguy>
(the "pcie-apple" stuff in u-boot/linux)
<amarioguy>
ditto the rc question for "port1", etc, does each P2P bridge have it's own ECAM region?
<roxfan2>
region count?
<roxfan2>
(no idea actually)
elros has joined #asahi-dev
compassion1785 has joined #asahi-dev
roxfan2 is now known as roxfan
elros has quit [Ping timeout: 480 seconds]
<eiln>
fyi it cannot be stateless as I'd hoped. we need a basic syntax parser to generate a 4K ish header. so far I've figured out height/width/stride. we also need to un-tile the output, and I'm really not sure if we want to do that in the kernel.
<eiln>
neither bit is in the moronic m3 blob, so we have to do it regardless
elros has joined #asahi-dev
chadmed has joined #asahi-dev
<nicolas17>
eiln: what will be the userland API to access it?
<jannau>
eiln: we can't recreate the header from the stateless structs? what information is in that header
<jannau>
nicolas17: v4l2
<nicolas17>
I suspect Apple accesses it from something with a lot more codec awareness
<jannau>
I don't follow. the v4l2 stateless decoder api requires quite a bit of codec knowledge from the API user
<nicolas17>
hm looks like AppleAVD.kext has H264 parsing (and it has lead to CVE-2022-22675)
eldriitch has quit [Quit: Leaving...]
roxfan has quit [Read error: Connection reset by peer]
<eiln>
jannau: I slightly misunderstood. the headers *are* chained (the second & onward headers do much less than the first), but it might be just an optimization. either way I can only decode one frame right now.
<nicolas17>
so we get hardware-accelerated decoding of HEIF photos? :3
eldriitch2 has joined #asahi-dev
<jannau>
eiln: if you have questions regarding video bitstreams feel free to ask
<jannau>
assuming you're looking at h264, the picture parameter set is usually much smaller than the sequence parameter set
<sven>
amarioguy: it’s ecam plus some config space for each port for irqs, clocks, etc.
<sven>
s/config space/apple (or more likely apple+synopsys) MMIO region/
<sven>
and iirc some global MMIO that just needs some tunables applied
<sven>
there’s also some region for the PHY but I believe m1n1 takes care of that
kallisti has joined #asahi-dev
<kallisti>
nicolas17: chadmed : No, the MBP 14" M2 variant is not supported. The installer returns `This device is not supported yet! Please check out the Asahi Linux Blog for updates on device support:` and the device is not in the list...
eldriitch2 has quit [Ping timeout: 480 seconds]
<chadmed>
kallisti: yeah forgot about the wifi chipset and gpu carveout...
feliciss has joined #asahi-dev
<eiln>
jannau: thanks :D
<eiln>
I started with h264 but there was some funny SPS/PPS stuff, like you mentioned. also scaling. I'm doing vp9 rn
<jannau>
kallisti: it is listed there but this conversation seems off-topic here. support for device is not ready yet
<kallisti>
jannau: Sorry about that.
<jannau>
eiln: I would be a good person for avd/ave but I have my hands full with dcp
ydalton has joined #asahi-dev
<eiln>
oh wait no it's something with the reference frame
<ydalton>
eiln: wow, i did not expect this
<ydalton>
what's up with that 0xb7000 big structure that is referenced by the mailbox messages?
<ydalton>
i remember i ran R's tracer script and i saw that
<sven>
jannau: just die what alyssa did and find a poor volunteer who takes over :D
<sven>
s/die/do/ (wtf iPhone autocorrect)
<ydalton>
i'd like to help but i know jack about anything :P
<ydalton>
eiln: did you use your own tracer script?
<Guest328>
that’s one of the indirection layers
<Guest328>
the last one iirc, it contains the pointers to actual video data
feliciss has quit [Quit: Leaving.]
<Guest328>
it’s big because it has some arrays of structs e.g. per-slice within a frame
<amarioguy>
sven: ah got it, thanks for the info
<ydalton>
R called it the frame params
<ydalton>
both the mailbox and the frame params contain the frame number of the stream
<ydalton>
i don't have my notes at hand
<ydalton>
but why apple, why the indirection
<eiln>
ydalton: in dart stream 1?
<eiln>
"tracer" it just mon.poll()'s periodically
<ydalton>
i mean like didn't you run the hypervisor and use some kind of tracer script?
elros has quit [Ping timeout: 480 seconds]
<ydalton>
i'm assuming that avd_dma_tunables_stages0 is that shit the macOS driver spams in MMIO
darkapex3 has quit [Remote host closed the connection]
darkapex3 has joined #asahi-dev
<ydalton>
Guest328: is this jamie?
ydalton has left #asahi-dev [ERC 5.4 (IRC client for GNU Emacs 28.2)]
<ChaosPrincess>
eiln: w.r.t ave_log.cfg, either that didnt end up working, or i have no idea where exactly to look for output. And im not sure how that even was supposed to work, since the scanf format that they are using is "LOG_ID_%d=%d%c"
elros has joined #asahi-dev
<eiln>
echo "1 2 hello" > /tmp/ave_log.cfg
<eiln>
> VTEncoderXPCService: (AppleVideoEncoder) 4319867810 64 AVE WARN: used log config file is at /tmp/ave_log.cfg 1694906712191446