| agneli_ | hey gnarface, any success? | 09:23 |
|---|---|---|
| gnarface | agneli_: didn't get a chance to today, gonna have to try for tomorrow | 09:30 |
| gnarface | if i don't get a chance to borrow that machine tomorrow morning i may have to just buckle down and install pulseaudio on a laptop or something | 09:31 |
| gnarface | i really don't like pulseaudio, and i really don't want it touching any of my stuff, so there's some twisted irony at being forced to use it more just to get away from it | 09:32 |
| gnarface | but i think i have a spare laptop i can use for it, i'm just being lazy | 09:33 |
| agneli | gnarface: build a vn | 10:22 |
| agneli | just for tests, no? | 10:22 |
| gnarface | a VM? hmm, i wonder if Steam would work in a VM... | 11:23 |
| gnarface | agneli: hmm, the busybox version of netstat may be useless... i think i'm seeing here about 25 open unix sockets but no network sockets? | 15:41 |
| gnarface | oh, wait no, there they are... stand by... | 15:42 |
| gnarface | agneli: check /msg | 15:44 |
| gnarface | only thing i can guess from that is audio and video must both be going to the same port, and "streaming_client" is the thing that deals with both | 15:47 |
| gnarface | hmm... now suddenly i'm wondering if it's possible i could just mknod the pcm devices | 16:07 |
| gnarface | i was thinking i'd have to build a kernel for it, but if it's really got alsa in there, maybe it's just a couple mknod commands from working normally with the alsa tools in devuan | 16:10 |
| agneli | gnarface: what about simply copying aplay? | 19:31 |
| agneli | from some 3.8.x based distro? | 19:32 |
| gnarface | agneli: i did do that, from devaun ascii. aplay, arecord, and speaker-test all say alsa isn't present, but suddenly i just thought of the fact only timer is in /dev/snd and there's no udev or systemd so maybe i just have to mknod some devices and they'd magically appear? | 20:31 |
| gnarface | just not sure what exact mknod parameters to use... | 20:31 |
| gnarface | (just read your /msg) | 20:33 |
| gnarface | that was the example with pulseaudio, but i wasn't streaming any music manually, it was just using the built-in automatic steam audio + video streaming, so yea i'm inferring they must both go through the other port (the ssh connection is just the terminal i'm using to get the info) | 20:34 |
| gnarface | i know that i can get netcat to work, i did successfully test it | 20:34 |
| gnarface | now i just need to figure out how to get aplay to work and my first guess is the /dev/snd nodes are missing, but my question is are they missing just because nothing created them, or are they missing from the driver code itself | 20:35 |
| gnarface | i debootstrapped devuan ascii onto a usb key and chrooted it, and the normal text tools i've tested seem to work fine, but alsa stuff is not | 20:36 |
| gnarface | and i think /proc/asound/cards said "--- no soundcards ---" or something like that | 20:37 |
| gnarface | but it's clearly got a soundcard | 20:37 |
| gnarface | sound comes out the HDMI port | 20:37 |
| gnarface | even when nothing is connected, there's sound in the UI, so it's definitely not just passing data through | 20:38 |
| gnarface | the 1-port thing does bring up the idea that there must be some way to get the steam host to pick up the audio and mix it into the video stream at that end, but the question there is how much of pulseaudio i'd have to re-implement to make it work | 20:40 |
| gnarface | are those mknod major/minor numbers the same for all devices? they're not, right? i can't just copy the ones from my desktop for example, can i? | 20:51 |
| gnarface | they'd be soundcard model specific, right? | 20:52 |
| agneli | you just want to do the mknod, right? :) | 20:59 |
| agneli | you believe it is not worth trying the sockets this application uses? | 21:00 |
| agneli | u have root there? | 21:00 |
| gnarface | yea, i have root but the application uses exactly one socket and it's in use already, i can't see what i could do with that | 21:03 |
| gnarface | i'm assuming it must encode the audio into a single stream with the video at the host end | 21:04 |
| gnarface | so i was just trying to route around it since i figured it wasn't useful | 21:04 |
| gnarface | though if it had some docs that'd be great... | 21:04 |
| gnarface | but the original plan was to just try to route around it temporarily with netcat, yea, so if i can find a shorter path to that than building a whole kernel that'd be fine | 21:05 |
| agneli | one? | 21:11 |
| agneli | I can see like 4 or 5 sockets | 21:11 |
| agneli | yes 5 different idnoes | 21:12 |
| gnarface | but only one of them is is streaming_client | 21:12 |
| gnarface | i think the others are all to do with stupid networking features i'm not using (wifi and bluetooth) | 21:13 |
| gnarface | the only one that appears after i connect the host is that one streaming_client instance | 21:13 |
| agneli | hmm i 5 different inodes for the pid 2355 | 21:13 |
| gnarface | oh, hmm... | 21:13 |
| gnarface | well the command line has a bunch of config files loading | 21:14 |
| gnarface | maybe it's those? | 21:14 |
| agneli | and one udp network connection yes | 21:14 |
| agneli | config files you would see in /proc/2355/fd | 21:15 |
| agneli | I am sure these soectes are listed there separately | 21:15 |
| gnarface | hmm... | 21:15 |
| gnarface | i'll have to check it later | 21:15 |
| agneli | :) | 21:16 |
| gnarface | thanks | 21:16 |
| agneli | yw | 21:16 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!