mrmidi.bsky.social

That was a journey. Single dev, complete FireWire stack from scratch: OHCI, isoch, async. DriverKit+PCIDriverKit+AudioDriverKit Bonus: FireWire audio - some of my older code was reused from previous experiments. @sakamocchi.jp - thanks for help :)

I've poked AppleFWAudio.kext in disassembler regarding mLAN - close to none useful info. I suppose there is no sense digging it w/o specs at all. But I still missing ohci 1.2 draft and some ta1394 draft specs: i've spent couple of months to understand that control word was wrong for 1.2 for AT reqs

@sakamocchi.jp have you ever thought on reversing mLAN protocol?
source: Implementing Digital Audio Devices for
the IEEE 1394 High Performance Serial Bus
Bob Moses
PAVO, Inc.
I'm just curious — it's not call to action :)

I hope I’ve responded like an adult :) Never imagined I’d contribute to the Linux kernel and receive such a considerate review. With macOS dropping FW, I started exploring OHCI adapters, AudioDriverKit & DriverKit. Linux drivers are an invaluable source of truth I deeply respect your expertise.

Tired of recompiling just to tweak a parameter.
Moving to a GUI — time to stop suffering.

GitHub: github.com/mrmidi/FireW...
Discord: discord.gg/c82rmSEEPY

I did it. Apogee Duet FireWire on macOS sequoia (arm) without AppleFWAudio.kext

Well, got myself FireWire packet sniffer :)

It’s such a joy to have someone as knowledgeable as you to turn to when I’m completely stuck. You always help, and I truly appreciate it. Thank you!

Yay! I found it! There are Get Vendor Dependent and Set Vendor Dependent commands starting with bytes 0x00 or 0x01. Thank you so much — your help has been invaluable once again :)
By the way, opcode 0x08 is Mics Grouped. I’d be happy to share more findings if you’re interested, or I can make a PR

I've skipped 0x00ff0000 part. I don't have an idea where it comes from.

I know, that 0x00 is vendor-dependent, the rest I've got by exploring your rust code. 0x09 - OUT_MUTE. To check my guess I've added some lines of code to tests.
As for the command 0x00ff0000 03db5043 4d0980ff, it’s being rejected:
0a ff 00 00 03 db 50 43 4d 09 80 ff. At least that’s something new :)

I referred to the AV/C Digital Interface Command Set General Specification Version 4.2, but unfortunately, vendor-dependent commands are covered in only one brief paragraph, which isn’t very helpful.
Could you please advise me on where to look or how to proceed?

When executing these commands, I always get the first byte of the response as 0x08 (not implemented). I suspect I might be missing some kind of initialization step or specific setup. I’ve tried various commands, including those from the snd-firewire-ctl-services tests, but so far, without success.

Thank you again for your previous help — thanks to your guidance, I managed to figure out isochronous streams and even decode AM824 (basic). However, I am now completely stuck with vendor-dependent AV/C commands.

Thanks again. My idea is to create a open source driver for obsolete devices on macOS in user space since apple removed their FireWire audio kext. I think that is possible to create such tools using audiodriverkit framework. Your code is big inspiration for me and a starting point also.

I just followed you and saw that you’re seeking a kofola :)
Thank you kindly for your help

@sakamocchi.jp Takeshi-san, Apogee Duet FireWire (OXFW971 ASIC) on macOSのisochronous transmissionを有効にしようとしています。初期化、機能検出、インターフェイス作成は完了しました。フォーマットとサンプルレートの設定後、伝送を開始するために追加の手順が必要ですか？ ちなみに、私はČRに住んでいるので、Kofolaを送ることができます :)

Кошка Тиса запрещает мне программировать безальясинговые пилообразные осцилляторы в c++

Тут тихо но уютно

О доброте