qyliss changed the topic of #spectrum to: A compartmentalized operating system | https://spectrum-os.org/ | Logs: https://logs.spectrum-os.org/spectrum/
<cole-h> You know what would be nice? A cachix cache for {chromiumOS,spectrum}Packages :P
* cole-h is still compiling chromiumOSPackages.linux
cole-h has quit [Quit: Goodbye]
c4rc4s_ has joined #spectrum
IdleBot_85f8451c has joined #spectrum
c4rc4s has quit [*.net *.split]
FireFly has quit [*.net *.split]
IdleBot_3d9d0c82 has quit [*.net *.split]
c4rc4s_ is now known as c4rc4s
FireFly has joined #spectrum
tilpner has joined #spectrum
tilpner_ has joined #spectrum
tilpner has quit [Read error: Connection reset by peer]
tilpner_ is now known as tilpner
tilpner_ has joined #spectrum
tilpner has quit [Ping timeout: 246 seconds]
tilpner_ is now known as tilpner
cole-h has joined #spectrum
<qyliss> Planning on streaming from 18:00 UTC
<leah2> yay, evening programme :3
<cole-h> Woot!
MichaelRaskin has quit [Read error: Connection reset by peer]
MichaelRaskin has joined #spectrum
<qyliss> Okay, I think I am live!
<tazjin> I'm finishing up some food, will tune in soon
<cole-h> Hey, ho! Let's go!
<Profpatsch> where’s live?
<cole-h> mpv rtmp://hyperion.kookie.space:11011/alyssa.tv
<Profpatsch> fancy
<cole-h> Nice meeting you here :^)
<qyliss> I still haven't figured out how to fix the audio, so I'll only be coming to your left ear unless you do something on your end I'm afraid
<qyliss> Somewhere in the logs there's the flags to mpv to fix the audio
<qyliss> If somebody wanted to find and post that that'd be great
<cole-h> = af toggle "lavfi=[pan=1c|c0=0.5*c0+0.5*c1]" ; show-text "Audio mix set to Mono"
<Profpatsch> I’m getting [ffmpeg] tcp: Connection to tcp://hyperion.kookie.space:11011 failed: Connection timed out
<qyliss> Could somebody please confirm what they see on the stream?
<cole-h> I see GNU Emacs :)
<qyliss> Lovely
<qyliss> Profpatsch: did you include the rtmp:// ?
<Irenes[m]> I'm there
<cole-h> Profpatsch: With mpv? Should be rtmp://
* FireFly opens the stream
<cole-h> I can hear you!
<FireFly> I can hear you, on my left ear :p
<cole-h> In one ear! (before hitting my keybind)
<cole-h> :D
<FireFly> though I think there's a flag for that
<Profpatsch> It is, hm
<FireFly> what's the mpv keybind? :p
<cole-h> = af toggle "lavfi=[pan=1c|c0=0.5*c0+0.5*c1]" ; show-text "Audio mix set to Mono"
<FireFly> o
<FireFly> no worries
<tazjin> qyliss: s/assi/asi/
<FireFly> cole-h++, ty :D
<cole-h> :P No bot here
* FireFly put that in $XDG_CONFIG_HOME/mpv/input.conf
<Profpatsch> hold your horses, I’m installing vlc
<leah2> owo
<Profpatsch> hrm, cannot connect as well.
<Profpatsch> This is a v4/v6 thing, isn’t it
<FireFly> hm
<leah2> v6 works fine :p
<Irenes[m]> for users listening through VLC, if you choose Audio > Stereo Mode > Mono it will map it to both ears on your end
<leah2> Profpatsch: use the raw ip i guess
<cole-h> Quick question: will you be running through how to actually run crosvm and test its functionality (no matter how little)?
<cole-h> `cargo run -- run --disable-sandbox result/bzImage` does indeed start stuff up, but it immediately closes for me
<cole-h> (where result/bzImage is from chromiumOSPackages.linux)
<Profpatsch> oh wow, it works if I tunnel trough my wireguard interface oOoO
<Profpatsch> wtf
<Profpatsch> My provider must be trippin again
<Profpatsch> deep packet inspection? QoS? who knows!
<cole-h> `nix-shell -p spectrumPackages.spectrum-vm` and away we go (while it builds)!
<cole-h> Got it, that was gonna be my next question. Does `spectrum-vm` create a rootfs?
<Profpatsch> you can also give --af=lavfi='[pan=1c|c0=0.5*c0+0.5*c1]' on the mpv command line
<cole-h> Hah, I remember you told me to do that last time I asked, too... Oops.
<FireFly> :D
<cole-h> :)
<FireFly> cole-h: I didn't pick it up last time, I'm glad you asked again
<cole-h> (Last time being ~yesterday lol)
<FireFly> oh ah
<MichaelRaskin> I did run CrosVM, but put it into my own sandboxing and called it a day
<cole-h> (When I was procastinating writing my final)
<cole-h> So, why is it an issue when running from the repo, again?
<MichaelRaskin> Nix package patches the path
<cole-h> Ah, it looks for seccomp stuff in /usr/share or whatever by default, right?
<cole-h> And since we don't have /usr/share... lol
<cole-h> Confirmed.
* cole-h is on Arch
<cole-h> OK, cool, thanks for the enlightenment.
<cole-h> Oh, so getopt is why we don't want to just tack on extra args and call it a day
<cole-h> tbh `-a` is a decent idea. Just make it a quoted string, though (then we'd have to split it, but...)
<cole-h> I was thinking e.g.: spectrum-vm -a '--disable-sandbox --trackpad 1:2:3'
<FireFly> using -a and adding in the -- sounds pretty decent for command-line options
<cole-h> Fair enough.
<FireFly> cole-h: I think that has the problem of starting with double-dash still
<cole-h> Yeah, I like double double-dash tbh.
<MichaelRaskin> I think some tool went to triple-dashes
<Irenes[m]> the notation with double dash-dash precludes crosvm from ever using dash-dash for its own purposes
<cole-h> LOL
<Irenes[m]> which is fine if it doesn't...
<FireFly> heh
<cole-h> Can we be smart and notice that "if the arg after -- doesn't start with a single- or double-dash, treat it as a command to send to cros"?
<cole-h> So we don't have to `spectrum-vm -- -- asdf`
<Profpatsch> execline blocks?
<Profpatsch> heh
<cole-h> lol
* cole-h is rebuilding his cros kernel.
<cole-h> Well, that's nice, lol. No need to workaround something now :D
* cole-h has to resist the temptation to fix those warnings.
<cole-h> The real question is: which ones are upstream, and which one are yours lol
<cole-h> More shortlog lines >:D
<cole-h> I only asked because master has warnings too (which may be yours lol)
<cole-h> Oh, great. Should I do master first and then let you merge that into interguest, or are the branches mostly unrelated?
<cole-h> OK, then I'll do master-specific stuff first and then interguest afterwards.
<cole-h> \
<cole-h> Been so many times in me debugging nixpkgs stuff that I've forgotten that and then failed the build after 5 minutes...
<Irenes[m]> I think I have the general idea
<cole-h> CROSVM_OPTS seems like an env var tbh
<cole-h> Clear enough
<cole-h> I would personally do `CROSVM_OPTIONS`
<cole-h> over OPTS
<cole-h> Calculated.
<FireFly> :D
<FireFly> nice
<cole-h> 100% calculated.
<cole-h> Doesn't help that 50% of that line is the actual package lol
<Profpatsch> haha, my color scheme is very similar, using birds-of-paradise here
<cole-h> nix-shell -p spectrumPackages.spectrum-vm --run 'spectrum-vm -C target/debug/crosvm -- --disable-sandbox'
<cole-h> :D
<cole-h> It's certainly an improvement on docbook, IMO x)
<cole-h> Don't you already have an example block above? Like with the tree dir structure?
<cole-h> Ah, listing != example
<cole-h> Because example will label with example or something, right?
<FireFly> makes sense to use what feels semantically appropriate, and then instruct the tools to render that as you want it, to me
<FireFly> also, looking at that help you have open, I guess maybe the tree(1) output should be a 'literal' (verbatim output), since it sounds like 'listing, source' specifies being for input rather than output
<FireFly> (but, I don't really know asciidoc, and either way that might be unnecessarily nitpicky)
<cole-h> Does adoc have a way to do small snippets of code, e.g. Markdown's backquotes?
<cole-h> Good, how are you? :)
<Irenes[m]> I'm okay
<tazjin> nominal pandemic mood
<Irenes[m]> got to do a thing though, back later I hope
<MichaelRaskin> `+variable+` seems to be «monospaced and no formatting»
<Irenes[m]> this was nice
<cole-h> MichaelRaskin: You to it just before I did :(
<cole-h> s/to/got to/
<cole-h> Note that the backticks is required
<cole-h> s/is/are/
<cole-h> Backtick-plus disables formatting
<cole-h> You can do markdown-like formatting with just backticks, but you can do `asdf *sadf*` which will embolden `sadf`
<cole-h> Inside the monospace snippet
<MichaelRaskin> Monospace + no formatting is what I would expect from a code «piece»
<cole-h> `+ just means `+*asdf*+` will appear as `*asdf*` monospaced
<cole-h> Whereas `*asdf*` will show up as bold `asdf`
<cole-h> Btw, a literal block (probably what you want for the tree) is `....` instead of `----`
<cole-h> And also, I found this playground for asciidoc: https://asciidoclive.com/
<cole-h> Side-by-side of source adoc + output html
<cole-h> (.... was meant for the tree example)
<cole-h> :P
<cole-h> Yep, that's good.
<cole-h> Would it be worth it to prepend the commands with `$`?
<cole-h> Fair enough. I guess that's one thing docbook succeeds at: separate the prompt from the actual command
<cole-h> No big deal, indeed
<cole-h> `doas cargo build` lol
<cole-h> Oh hey, kernel finally finished building
<cole-h> lol
<FireFly> I think so
<FireFly> hm
<FireFly> you might have to search for 'Jonas Höglund', not sure
<MichaelRaskin> Do we have IRC handles in maintainer list? Maybe we should
<FireFly> oh, you could also git blame the maintainers list for the FireyFly entry I guess (so-named because github handle)
<cole-h> That would be nice, tbh. It took me a while to relate you and your GH handle :D
<MichaelRaskin> Well, technically Firefly would match in the email domain
<FireFly> oh point
<MichaelRaskin> And I even have my name in the GH bio!
<FireFly> lol
<FireFly> yeah sure, tha might be easier
<FireFly> that*
<cole-h> MichaelRaskin: It's true. I just don't hover on GH handles much, and never had a reason to your profile :P
<MichaelRaskin> It was fun when I did not set display name in email client
* FireFly waiting for the spectrum-vm build to finish as well
<MichaelRaskin> Ah, GH _emails_ randomly use handles and names in different places
<FireFly> ...maybe blaming the maintainer list wasn't a great suggestion after all >.< whoops :p
<MichaelRaskin> What _blaming_ m-l can ever yield??
<MichaelRaskin> I mean, _searching_ is enough 100% of time, no?
<leah2> use tig!
<FireFly> you can use the firefly@ one
<MichaelRaskin> I am pretty sure I have been a committer for longer than maintainership records exist in files
<cole-h> `git log --grep='FireyFly'` shows the commit immediately :P
<cole-h> Maybe you were misspelling it?
<FireFly> oh yeah
<cole-h> OK, I must depart. Good luck with the rest of the stream (however long that may be) and thanks for the intro on how to actually run spectrum :D
* FireFly feels the time taken for finding their email isn't really proportional to the amount of help "looks like .... seems more semantically correct" provided :p
<cole-h> Cleanup commits are my specialty!
<FireFly> seeya, cole-h
<FireFly> \o/
* FireFly nods
<tazjin> sleep well :P
<FireFly> get some rest and get the things you wanted to get done done tomorrow :)
<MichaelRaskin> Good night.
<qyliss> Yes :)
<qyliss> Night!
<leah2> nini
<cole-h> Got crosvm to run! Only downside is the keyboard layout is most definitely not qwerty lol
<cole-h> qyliss: btw the warnings are present on Rust 1.43.0, not 1.42.0. I'll still fix 'em up, though (eventually).
<qyliss> cole-h: if they're Rust 1.43-specific I wouldn't bother. Google will fix those.
<cole-h> SGTM. No patches (just yet) for you then ;^)